DH Kenya Report

DISRUPTING
HARM IN
KENYA
Evidence on online child
sexual exploitation and abuse
Funded Implemented
by by
Suggested citation:
ECPAT, INTERPOL and UNICEF. (October, 2021). Disrupting Harm in Kenya: Evidence on online child
sexual exploitation and abuse. Global Partnership to End Violence against Children.
© ECPAT, End Violence Partnership, INTERPOL, UNICEF, 2021. Use of this publication is permitted
provided the source is acknowledged and that the publication is not used for commercial purposes.
Funding from the Global Partnership to End Violence against Children, through its Safe Online
2 initiative, does not constitute endorsement. Disrupting Harm in Kenya – Evidence on online child sexual exploitation and abuse
CONTENTS
Foreword 4
Executive Summary 6
Disrupting Harm Methods 11
About online child sexual exploitation and abuse 15
About Kenya – demographics and internet usage 17
Overview of legislation and policy 21
1. Children Online in Kenya 22
1.1 Internet access and barriers 23

1.2 Children’s activities online 26
1.3 Perceptions and experiences of risky online activities 27
2. Online Child Sexual Exploitation and Abuse in Kenya 36
2.1 Law enforcement data 38

2.2 Children's experiences of child sexual exploitation and abuse in Kenya 44
2.3 Other experiences of children that may be linked to OCSEA 60
2.4 Insights about victims and offenders from known OCSEA and CSEA cases 65
2.5 Barriers to children speaking to adults about OCSEA 66
3. Responding to Online Child Sexual Exploitation and Abuse in Kenya 70
3.1 Formal reporting mechanisms 71

3.2 Law enforcement response 73
3.3 Obtaining justice and access to remedies 80
3.4 Coordination and collaboration 86
4. How to Disrupt Harm in Kenya 92
Acknowledgements 101
Disrupting Harm in Kenya – Evidence on online child sexual exploitation and abuse 3
FOREWORD
Our online lives are advancing constantly. The internet and

rapidly evolving digital communication tools are bringing
people everywhere closer together. Children are increasingly
conversant with and dependent on these technologies, and
the COVID-19 pandemic has accelerated the shift online of
many aspects of children’s lives.
The internet can be a powerful tool for children to connect, explore, learn
and engage in creative and empowering ways. The importance of the digital
environment to children’s lives and rights has been emphasised by the United
Nations’ Committee on the Rights of the Child in General Comment No. 25
adopted in 2021. The General Comment also stresses the fact that spending time
online inevitably brings unacceptable risks and threats of harm, some of which
children also encounter in other settings and some of which are unique to the
online context.
One of the risks is the misuse of the internet and digital technologies for the
purpose of child sexual exploitation and abuse. Online grooming, sharing of
child sexual abuse material and live-streaming of child abuse are crimes against
children that need an urgent, multi-sectoral and global response. These crimes
are usually captured in permanent records in the form of digital images or
videos, and are perpetually reshared online, victimising children over and over
again. As risks of harm continue to evolve and grow exponentially, prevention
and protection have become more difficult for governments, public officials and
providers of public services to children, but also for parents and caregivers trying
to keep-up with their children’s use of technology.
With progress being made towards universal internet connectivity, it is ever-more

pressing to invest in children’s safety and protection online. Governments around
the world are increasingly acknowledging the threat of online child sexual
exploitation and abuse, and some countries have taken steps to introduce the
necessary legislation and put preventive measures in place. At the same time,
the pressure is mounting on the technology industry to put the safety of children
at the heart of design and development processes, rather than treating it as an
afterthought. Such safety by design must be informed by evidence; Disrupting
Harm makes a significant contribution to that evidence.
4 Disrupting Harm in Kenya – Evidence on online child sexual exploitation and abuse
The Global Partnership to End Violence against Children, through its Safe Online initiative,
invested US$ seven million in the Disrupting Harm project. Disrupting Harm uses a holistic
and innovative methodology and approach to conducting comprehensive assessments
of the context, threats and children’s perspectives on online child sexual exploitation and
abuse. This unprecedented project draws on the research expertise of ECPAT, INTERPOL,
UNICEF Office of Research – Innocenti, and their networks. The three global partners were
supported by ECPAT member organisations, the INTERPOL National Central Bureaus and
the UNICEF Country and Regional Offices. It is intended that the now developed and
tested methodology is applied to additional countries around the world.
Disrupting Harm represents the most comprehensive and large-scale research project
ever undertaken on online child sexual exploitation and abuse at a national level and has
resulted in 13 country reports and two regional reports. It provides the comprehensive
evidence of the risks children face online, how they develop, how they interlink with
other forms of violence and what we can do to prevent them.
The findings will serve governments, industry, policy makers, and communities to take the
right measures to ensure the internet is safe for children. This includes informing national
prevention and response strategies, expanding the reach of Disrupting Harm to other
countries and regions, and building new data and knowledge partnerships around it.
Disrupting harm to children is everyone’s responsibility.
Dr Howard Taylor
Executive Director
End Violence Partnership
EXECUTIVE SUMMARY
Funded by the Global Partnership to End Violence against Children, through its
Safe Online initiative, ECPAT, INTERPOL and UNICEF Office of Research – Innocenti
worked in partnership to design and implement a multifaceted research project
on online child sexual exploitation and abuse: Disrupting Harm. The research
was conducted in six Southeast Asian countries and seven Eastern and Southern
African countries, including Kenya. Data is synthesised from nine different research
activities to generate each national report which tells the story of the threat, the
response to that threat, and presents a clear way forward.
Perceptions of online risks Although these encounters tended to be positive for

The great majority of caregivers in Kenya are highly children in our survey, this remains a risky activity, the
concerned that their children will talk to people outcome of which can be highly variable. Children
online whom they have never met in person, or should therefore not meet with online contacts in
encounter sexual images on the internet. Caregivers person without taking some safety precautions like
generally use the internet less than their children, informing a trusted adult, only meeting in public
and their ability to guide them may be limited. places, and never meeting the person alone. Six
per cent (60 children) had shared naked images or
Meanwhile, two-thirds of internet-using children videos of themselves with other internet users in the
have not been taught about how to stay safe online. past year. While such images are most frequently
Their awareness of the risks varies. In the Disrupting shared voluntarily among peers and close friends,
Harm Kenya household survey of 1,014 internet- seven children had shared naked images as a result
using children age 12-17 and their caregivers, 14% of threats and six said they were pressured by
of the children had met someone face-to-face after their friends.
first encountering them online in the past year.
According to children, many of these encounters did Potential and actual instances of sexual
not result in immediate harm and most respondents exploitation and abuse
described being pleased about the outcome. As part of the household survey, children were
asked whether they had been subjected to a range
of potential and actual instances of online sexual
exploitation and abuse (OCSEA) within the past
Children are subjected to year. OCSEA refers to situations that involve digital or
communication technologies at some point during
these potential and actual the continuum of abuse or exploitation. OCSEA can
instances of sexual exploitation occur fully online or through a mix of online and in-
person interactions between offenders and children.
and abuse both online and
Potential instances of OCSEA included, for example,
offline. In Kenya, most children unwanted requests to talk about sex and unwanted
who have been exposed to requests for images showing their private parts.
Actual instances of OCSEA included, for example,
any manifestation of OCSEA being offered gifts in return for sexual images and
have also been subjected to being threatened or blackmailed online to engage
in sexual activities. The proportions of children who
physical, sexual or emotional said that these things had happened to them varied
violence in person. between 5% and 13%, depending on the question.
Most children who were subjected to possible
grooming attempts refused to do as asked. However,
a small proportion complied with unwanted
requests to talk about sex or share sexual images.
Seven percent of all children surveyed said sexual Some people hold the
images of them had been shared without their
permission in the past year. belief that sexual abuse that
Children are subjected to these potential and actual only happens in the online
instances of sexual exploitation and abuse both environment is not ‘real’
online and offline. In Kenya, most children who have
been exposed to any manifestation of OCSEA have abuse.
also been subjected to physical, sexual or emotional
violence in person.
Consistent with the evidence about violence against someone, confided mainly in their friends. Only a
children offline, persons already known to the child minority had told their caregivers or other adults
were responsible for most of the potential and actual and very few went to the police or spoke to a
instances of OCSEA disclosed by respondents of the social worker or helpline. The main reasons given
household survey. In many cases, these persons were for not disclosing were a lack of awareness about
other children. Even so, there were many instances where to go or whom to tell, feeling embarrassed
where the offenders were adults. These instances or ashamed or that it would be emotionally too
could be evidence of grooming of children with difficult to talk about it, and fear of getting in trouble.
a view to sexually abusing and exploiting them in Frontline workers attributed this to strong taboos
person and/or for the creation of child sexual abuse regarding sex in Kenya. As shown in the household
material (CSAM). People unknown to the child survey with children, these feelings of shame and
were involved in about one incident in four. This has embarrassment affect both girls and boys. The
significant implications for prevention and awareness criminalisation of homosexuality adds an additional
raising, as many awareness efforts focus primarily on layer of stigma that may contribute to silencing
the threat from strangers rather than people the child children who are abused by an offender of the
already knows. This should also be a consideration for same sex.
response systems, as it could be much more difficult
for victims to seek help if they are emotionally and/or Interviews were conducted with nine Kenyan girls
economically dependent on abusers. who had been sexually assaulted by offenders who
got to know them online. These girls revealed that
Disclosure of online sexual exploitation they were initially deceived by compliments, gifts
and abuse and promises and did not realise they were being
Many incidents of OCSEA go undisclosed and groomed into OCSEA. Later they were too ashamed
formally unreported. Approximately one-third of and fearful of stigma or repercussions to tell anybody.
the children surveyed who had been subjected to Eventually they went to individuals or organisations
OCSEA had told nobody, and those who did tell outside the family for support, or were obliged to tell
their stories because they had become pregnant.
While most caregivers say they would tell the police

if their children were sexually abused, others said
they would not report due to concern about negative
Many children in Kenya consequences, fear of not being treated properly
did not tell anyone the last and/or a belief that reporting would have no effect.
In our frontline workers’ survey, more than half
time they were subjected to of respondents said that cases of OCSEA are not
OCSEA. being reported because services are not trusted. In
addition, frontline workers said that many adults
may not know that certain types of behaviour online
EXECUTIVE SUMMARY
are serious and punishable offences. Some people plugged when the upcoming Children Bill 2021,
hold the belief that sexual abuse that happens in the currently before Parliament, is enacted.
online environment is not ‘real’ abuse.
Among the young people who were interviewed
Law enforcement data about their experiences in accessing justice
The numbers of OCSEA cases handled by the Kenyan after being victims of OCSEA, some had positive
police were not provided on a national level but rather experiences with the police but most were
from the specialised unit. The Anti Human Trafficking disappointed. For example, some children were
and Child Protection Unit (AHTCPU) of the Directorate not informed about their rights or questioned
of Criminal Investigations alone handled 3,160 OCSEA sensitively. Some complained of officers expressing
cases in 2018 and 4,133 in 2019. These are the reported harsh opinions and judging them. Most children
numbers and do not provide a complete picture and caregivers did not sense a determination to
of OCSEA prevalence. Frontline workers surveyed bring the offenders to justice. Some caregivers
indicated that the offenders are most commonly felt that they were left out of the process. Only
adult members of the community in which the child a minority of regular police stations have child
lives. Some limited evidence was discovered of links protection units and these may not be able to
between OCSEA and travel and tourism. retain trained staff. The vast majority of the frontline
workers we spoke to rated the law enforcement
Between 2017 and 2019, the Kenyan law enforcement officers’ awareness and response to OCSEA as ‘poor’
authorities received an average of 13,572 CyberTips or ‘fair’.
per year from globally popular online platforms
based largely in the United States via the U.S. Children, justice and social services
National Center for Missing and Exploited Children Similarly, most children and caregivers who had
(NCMEC). In 2020, the number was 14,434. Almost experiences of going through the formal justice
all of these reports concerned apparent cases of the system for an OCSEA case found it discomforting.
possession, manufacture and distribution of CSAM in Not all cases can be heard in a Children’s Court,
Kenya. While Facebook submitted 93% of the reports, where child-friendly procedures are favoured. The
numerous other electronic service providers also Victim Protection Act envisages that the dignity of
submitted reports, suggesting the misuse of a range victims be preserved, and that each victim should
of platforms by OCSEA offenders. Research using be treated in accordance with his or her age and be
Google Trends points to interest in CSAM in Kenya protected from secondary victimisation. However,
including image and video content depicting sexual there are no specific operating procedures for how
activity with and between teenagers, with children, to implement this in practice. Knowledge of OCSEA
and with babies. among justice professionals is limited. Despite
the efforts made to put children at ease, several
Investigating cases and treatment children we spoke to still had to give evidence
of victims in the justice system repeatedly and faced stigma and victim-blaming.
Findings from the capacity assessments of law Victim impact assessments were not always carried
enforcement and access to justice interviews revealed out. Although children are entitled to free legal aid
that often the police and prosecutors have difficulty and counselling, these services were not provided
knowing how to recognise, investigate and prosecute to most of the child survivors we spoke to. Where
OCSEA cases. This reflects both gaps in legislation provided, they are not always free.
and a lack of access to training on these issues. The
Computer Misuse and Cybercrimes Act defines CSAM Investigations and trials of OCSEA cases are often
and criminalises acts associated with it quite explicitly, drawn out, which delays justice and prevents child
while the Sexual Offences Act sets the age of sexual victims from moving on. Besides pressure on the
consent at 18. However, no specific references are courts and lack of expertise, another reason for this
made in either law to live-streaming, online grooming is the difficulty of obtaining evidence from internet
for sexual purposes or sexual extortion in the online service providers, either because they respond late
environment. These potential loopholes may be to court orders or because they do not store data
for a sufficiently long period. Safaricom, the largest International and domestic civil society organisations
mobile telecommunications operator in Kenya, has play a major part in responding to OCSEA in Kenya.
reportedly taken steps to expedite its response to They refer cases to the police and the courts and
requests for evidence. cooperate with the Department of Children’s
Services in the provision of services like shelter,
The Victim Protection Act provides victims of counselling and legal aid. They are also involved in
crimes, including OCSEA, with free access to awareness raising activities and in training the child
counselling, shelter and reintegration services. protection workforce. However, these efforts are not
The Department of Children’s Services under sufficient to address these crimes fully.
the Ministry of Labour and Social Protection
seeks to provide these services to OCSEA victims Current initiatives for children
with the support of civil society organisations. Interviews with government duty-bearers
However, duty-bearers interviewed indicated demonstrate that the Government of Kenya is aware
that the Department itself is understaffed and of the threat of OCSEA and the need for cooperation
the availability of services is limited, especially in and collaboration to counter it. It has established
rural areas. More than half of the frontline workers a National Technical Working Group on Child
we surveyed rated the availability and quality of Online Protection which brings together mandated
psychological services as either ‘poor’ or ‘fair’. government agencies, civil society organisations,
industry representatives and UN agencies.
The Anti-Human Trafficking & Child
Government agencies have also conducted various
Protection Unit (AHTCPU)
awareness raising and training initiatives, albeit with
The establishment of the AHTCPU under the limited reach so far.
Directorate of Criminal Investigations, first in
Nairobi and more recently in Mombasa, is a good Two national policies which touch on OCSEA are the
step towards improving responses to OCSEA. National Plan of Action Against Sexual Exploitation
The Unit investigates cases of OCSEA which are of Children in Kenya, 2018-2022 and the National
reported to it in various ways or referred to it by Information, Communications and Technology
regular police stations. The Unit staff are specialised Policy (2019). However, neither policy has been
in such cases and have close links with the widely disseminated publicly across all relevant
Department of Children’s Services. stakeholders..
The Unit can also advise regular police stations, Two more specific policies are under development.
take over cases from them and provide training to One is the National Plan of Action on online child
other professionals on handling cases of OCSEA. sexual exploitation and abuse, spearheaded
However, it has a very limited staff: there are by the Department of Children’s Services. The
currently just five officers dedicated to the National Plan of Action is in the final stages of
investigation of OCSEA cases. development and is anchored in the WePROTECT
Model National Response. The second is a National
International cooperation and civil society Strategy on Child Online Protection, which is led
Both the AHTCPU and the National Kenyan by the Communications Authority of Kenya and
Computer Incident Response Team cooperate will embody the International Telecommunications
with INTERPOL and other international partners Union Guidelines on Child Online Protection.
to prevent and respond to OCSEA. There are
working arrangements between the Kenyan law Key insights
enforcement authorities and companies like The report for Kenya concludes by highlighting five
Facebook and Google for obtaining information key insights from the research:
during investigations. Several international 1. Internet-using children in Kenya are subjected to
organisations have supported Kenya with OCSEA. According to children who were subjected
mentorship, training, and equipment in areas to OCSEA and frontline workers, most offenders
related to OCSEA.
are someone the child already knows. These 5. Important OCSEA-related legislation, policies and
crimes can happen while children spend time standards are not yet enacted in Kenya.
online or in person but involving technology.
The report ends with a detailed map for action to be
2. Many children in Kenya did not tell anyone the
taken by the government, by the law enforcement,
last time they were subjected to OCSEA. Children
justice and social services sectors and by those
tend to disclose to people they know rather than
working within them, by communities, teachers
reporting to a helpline or the police.
and caregivers, and by digital platforms and service
3. Among children who were subjected to OCSEA providers. These are too detailed to be recounted in
through social media, Facebook and WhatsApp the Executive Summary but can be found on page 92
were the most common platforms where this of this report.
occurred.
4. The law enforcement, justice and social support
systems have inadequate awareness, capacity and
resources to respond to cases of OCSEA.
DISRUPTING HARM METHODS
As with all the settings in which children live and grow, the online environment
may expose them to risks of sexual exploitation and abuse. Yet, the scarcity of the
available evidence makes it difficult to grasp the nature of the harm caused or to
make constructive recommendations for governments’ approaches to prevention
and response. Informed by the 2018 WeProtect Global Alliance Threat Assessment1
the Global Partnership to End Violence against Children, through its Safe Online
initiative, decided to invest in research to strengthen the evidence base – with a
particular focus on 13 countries across Eastern and Southern Africa and
Southeast Asia.
The countries of focus in the Eastern and Southern Summary of methods used by ECPAT in Kenya
Africa region are Ethiopia, Kenya, Mozambique, Government duty-bearer interviews
Namibia, South Africa, Tanzania, and Uganda. The Twelve semi-structured interviews were conducted
countries of focus in the Southeast Asian region between May 2020 and July 2020 with a total of 16
are Cambodia, Indonesia, Malaysia, the Philippines, senior national duty-bearers4 with mandates that
Thailand, and Vietnam. include OCSEA. Due to the COVID-19 pandemic,
Extensive data collection for nine unique research some interviews were conducted in person and
activities took place in Kenya from early 2020 some virtually. More information on the methodology
through to early 2021 and focused on the three-year can be found here, while the preliminary report of
period of 2017-2019. During an extensive analysis this data can be found here. Attributions to data
phase, the data from all the research activities were from these respondents have ID numbers beginning
triangulated and a series of 13 country reports were with RA1 throughout the report.5
developed. Analysis for Kenya was finalised in May Analysis of non-law enforcement data and
2021. Using the same methodology in all 13 countries consultations
also allows for cross-country comparisons, which A range of non-law enforcement stakeholders have
will be presented in the two regional reports in data and insight on the nature and scale of OCSEA.
the series. The desired outcome of this report is to Data were obtained from INHOPE, the Internet
provide a baseline and evidence for Kenyan policy Watch Foundation and Child Helpline International
makers to tackle online child sexual exploitation and (CHI). Qualitative insight was provided by a number
abuse and strengthen victim support. In addition, of global technology platforms. Where relevant, this
the findings and advised next steps are expected to information supplements the analysis contributed by
have relevance for a broader global audience. The INTERPOL.
recommendations made in the report are aligned
with the WeProtect Model National Response2 and Frontline social service providers’ survey
contribute to the 2030 Agenda for Sustainable A convenience sample of 50 client-facing frontline
Development.3 workers in Kenya – such as outreach youth workers,
social workers, case managers, psychologists, and
some health and legal professionals directly working
1. WeProtect Global Alliance (2018). Global Threat Assessment 2018: Global Threat Assessment 2018: Working together to end the sexual exploitation
of children online. London: WeProtect Global Alliance.
2. WeProtect Global Alliance (2016). Preventing and Tackling Child Sexual Exploitation and Abuse: A model national response.
3. United Nations. (n.d.) Sustainable Development Goals. See: Goals 5.2, 8.7 and 16.2.
4. In this instance, duty-bearers are defined as those who hold specific responsibilities for responding to the risks of OCSEA at a national level.
Participants represented: the Communications Authority of Kenya, the National Council of Children’s Services, UNICEF Kenya Country Office, the
NCAJ Special Task Force on Children Matters, the Kenya Film and Classification Board, the Office of the Director of Public Prosecutions, the Child
Online Protection Unit, the Department of Children’s Services, the Ministry of Labour and Social Protection, the Kenya Law Reform Commission, the
Kenya Institute of Curriculum Development and the National KE-CIRT/CC.
5. The format RA1-KY-01-A is used for IDs. ‘RA1’ indicates the research activity, ‘KY’ denotes Kenya, ‘01’ is the participant number and ‘A’ indicates the
participant when interviews included more than one person.
with children’s cases – participated in a survey be found here. Attributions to data from these
administered online during April and May of 2020. respondents have ID numbers beginning with RA4
This research activity aimed to explore the scope throughout the report. Note that the suffix ‘justice’
and context of OCSEA as it is observed by those is also included in the ID numbers to indicate
working the social support front line to prevent it and interviews with justice professionals.
respond to it. More information on the methodology
can be found here, while the preliminary summary Literature review and legal analysis
report of this data can be found here. Attributions A literature review was undertaken to inform the
to data from these respondents have ID numbers research teams prior to primary data collection. A
beginning with RA3 throughout the report. comprehensive analysis of the legislation, policy and
systems addressing OCSEA in Kenya was conducted
Access to Justice interviews with OCSEA victims and finalised in June 2020. More information on the
and their caregivers methodology can be found here, while the full report
Ten interviews were conducted between June and on the legal analysis can be found here.
August of 2020 with children (all girls) aged between
15 and 18 years who had accessed the legal system Conversations with OCSEA survivors
for OCSEA cases. The girls’ caregivers were also Unstructured, one-on-one conversations led by
interviewed. The children and caregivers decided trauma-informed expert practitioners were arranged
themselves whether to be interviewed separately with 33 young survivors of OCSEA in five of the
or jointly. Only one child chose to be interviewed in Disrupting Harm countries (nine girls in Kenya, five
the presence of her caregiver. This research activity boys and seven girls in Cambodia, seven girls in
aimed to provide a better understanding of how and Namibia, four girls in Malaysia and one boy in South
to what extent child victims of OCSEA can access Africa). Participants were aged between 16 and 24
justice and remedies in Kenya. Despite deliberate but had all been subjected to OCSEA as children. The
efforts to identify males, the study in Kenya was survivor conversations were analysed collectively for
unable to identify any male children who had been all countries and lessons are weaved through all the
through the legal system. The female participants national reports. The Kenya report presents data from
for this activity came from seven of the 47 counties the nine Kenyan female survivors. More information
in Kenya – namely Migori, Nairobi, Eldoret, Meru, on the methodology can be found here. The report
Makueni, Nakuru and Mombasa. More information presenting the analysis of all 33 survivor conversations
on the methodology can be found here, while will be released separately in late 2021. Attributions
the preliminary summary report of this data can to data from these respondents have ID numbers
be found here. Attributions to data from these beginning with RA5.
respondents have ID numbers beginning with RA4 Summary of methods used in Kenya
throughout the report. Note that the suffix ‘child’ by INTERPOL
or ‘caregiver’ is also included in the ID numbers to
Quantitative case data analysis
indicate interviews with children and caregivers.
Data was sought on cases related to OCSEA from law
Access to Justice interviews with justice enforcement authorities via the INTERPOL National
professionals Central Bureau in each country. Data were also
Eleven semi-structured interviews were conducted obtained from the mandated reports of U.S. based
with twelve criminal justice professionals in June technology companies to the National Center for
and July 2020. The sample included State and Missing and Exploited Children (NCMEC) and from
non-State respondents who had experience a number of other partner organisations with a view
with OCSEA criminal cases.6 More information to deepening the understanding of relevant offences
on the methodology can be found here, while committed in the country, offender and victim
the preliminary summary report of the data can behaviour, crime enablers and vulnerabilities. Crime
data was analysed for the three years from 2017 to 2019.
6. The interviewees comprised one lawyer, one prosecutor, one magistrate, one representative from the Department of Children’s Services, three
police officers (two representing the AHTCPU and one who was a former investigator with the Cybercrime Unit), three counsellors (one from the
International Justice Mission, one formerly with Childline Kenya and the other formerly with CRADLE), and two child rights advocates involved in
outreach activities on child online protection – one from Watoto Watch Network and the other from Mtoto News.
Qualitative capacity assessments The sampling followed a three-stage random
In addition to seeking data on OCSEA cases, probability clustered sample design. At the first
INTERPOL requested data on the capacity of national stage, 100 primary sampling units (PSUs) were
law enforcement authorities to respond to this type selected. The PSU list was provided by the Kenya
of crime and interviewed serving officers. Particular National Bureau of Statistics (KNBS). At the second
emphasis was placed on human resources, access stage, interviewers randomly selected addresses
to specialist equipment and training, investigative in the field using random walk procedures and
procedures, the use of tools for international attempted contact at the selected addresses to
cooperation, achievements and challenges. screen for members of the survey population using
Attributions to data from this activity have ID a screening question developed for this purpose. At
numbers beginning with RA8 throughout the report. the third stage, individuals (children and caregivers)
were selected within each eligible household using
More information on INTERPOL’s methodologies can random methods.
be found here.
In every household visited we attempted to collect
Summary of methods used in Kenya by data on the number of 12–17-year-old children in the
UNICEF Office of Research – Innocenti household, their gender, and whether they had used
Household survey of internet-using children and the internet in the past three months. This allowed
their caregivers us to estimate internet penetration rates for all
In order to understand children’s use of the internet, 12–17-year-old children in Kenya.
the risks and opportunities they face online,
specifically OCSEA, a nationally representative The fieldwork took place between 21 December 2020
household survey was conducted with 1,014 internet- and 19 January 2021. Data collection was carried
using children. The target population for the survey out by IPSOS MORI on behalf of UNICEF Office of
was children aged 12-17 in Kenya who have used the Research – Innocenti.
internet in the three months before the interview.
A more detailed explanation of the methodological
Additionally, one parent or caregiver of each child was
approach and the specific methods used for the
interviewed. The interviews were conducted in person.
analysis of the household survey data can be found
To achieve a nationally representative random here.
sample, the survey used random probability
Ethical Approval
sampling with national coverage. In Kenya, fieldwork
coverage was 100%. Coverage is defined as the The ECPAT and UNICEF research components were
proportion of the total population that had a chance reviewed and approved by AMREF Health Africa –
of being included in the survey sample – meaning Science and Ethical Review Committee. UNICEF
that the fieldwork would cover the area where they also obtained required research approval for the
live if sampled. This means that all eight provinces of household survey from the National Commission
Kenya (Central, Coast, Eastern, Nairobi, North Eastern, for Science, Technology and Innovation (NACOSTI).
Nyanza, Rift Valley and Western) were represented ECPAT and UNICEF’s protocols were also reviewed
in our sample. Although in recent years counties and approved by HML Institutional Review Board.
have been used in preference to provinces in official INTERPOL has assessed the threat and the capacity
classifications, provinces were used to determine of law enforcement on countering the threat.
the proportional allocation of PSUs (stratification) Both assessments entailed interviews with law
in Kenya given the number of counties was too enforcement in relevant units and national agencies
many for this purpose. However, all counties had an dealing with OCSEA. Similarly to UNICEF, INTERPOL
equal chance of selection into the sample and the obtained NACOSTI approval. The team has taken an
population is representative at the national level. online course on Responsible Conduct of Research
from the Collaborative Institutional Training Initiative
and followed the INTERPOL’s Code of Conduct.
National Consultation were asked to comment on the Disrupting Harm

A national consultation took place on 8 June 2021. recommendations with the objective to ensure that
Government and non-governmental organisations the recommendations were relevant for the Kenyan
context.
Figure 1: Disrupting Harm methods in Kenya
Survivor conversations
n=9
Access to Law
justice enforcement
interviews capacity
Household with assessment
PHASE 2
survey data professionals n=9

from children n=11
and parents Frontline
n=1,014 service
providers’
survey
n=50
Access to
Government justice Non-law Country
duty-bearer interviews enforcement threat
Interviews with children data assessment
n=12 n=10
Desk review of relevant documents

PHASE 1
Legal analysis
ABOUT ONLINE CHILD SEXUAL EXPLOITATION AND ABUSE
Child sexual abuse refers to various sexual activities perpetrated on children

(persons under 18), regardless of whether the children are aware that what is
happening to them is neither normal nor acceptable. It can be committed by
adults or peers and usually involves an individual or group taking advantage of an
imbalance of power. It can be committed without explicit force, with offenders
frequently using authority, power, manipulation or deception.7
Child sexual exploitation involves the same abusive Any characterisation of OCSEA must recognise that the
actions. However, an additional element of exchange boundaries between online and offline behaviour and
for something (e.g., money, shelter, material goods, actions are increasingly blurred9 and that responses
immaterial things like protection, or even the mere need to consider the whole spectrum of activities
promise of such) must also be present.8 in which digital technologies may play a part. This
characterisation is particularly important to keep in
mind as children increasingly see their online and offline
Online child sexual exploitation and worlds as entwined and simultaneous.10
abuse (OCSEA) refers to situations involving
digital, internet and communication For Disrupting Harm, OCSEA was defined specifically
technologies at some point during the to include child sexual exploitation and abuse that
continuum of abuse or exploitation. OCSEA can involves:
occur fully online or through a mix of online
• Production, possession or sharing of child sexual
and in-person interactions between offenders
abuse material (CSAM): Photos, videos, audios or
and children.
other recordings, or any other representation of real
or digitally generated child sexual abuse or sexual
parts of a child for primarily sexual purposes.11
Differentiating between ‘online’ and ‘offline’ child sexual
• Live-streaming of child sexual abuse: Child sexual
exploitation and abuse as two completely separate
abuse that is perpetrated and viewed simultaneously
forms of violence does not help us to understand,
in real-time via communication tools, video
prevent or respond to the issue, nor is it the intention of
conferencing tools and/or chat applications. In most
Disrupting Harm to make such a distinction. Children
cases, the offender requesting the abuse in exchange
can be abused or exploited while they spend time in
for payment or other material benefits is physically
the digital environment, but equally, offenders can use
in a different location from the child(ren) and the
digital technology to document and share images of in-
facilitators of the abuse.
person abuse and exploitation.
• Online grooming of children for sexual purposes:
Disrupting Harm also focuses on how technology Engagement with a child via technology with the
facilitates child sexual exploitation and abuse and intent of sexually abusing or exploiting the child.
contributes much-needed evidence to understand While international legal instruments12 criminalising
the role digital technology plays in perpetrating grooming indicate that this must take place with
sexual violence against children. intent to meet the child in person, it has become
7. Interagency Working Group on Sexual Exploitation of Children. (2016). Terminology Guidelines for the Protection of Children from Sexual
Exploitation and Sexual Abuse. Bangkok: ECPAT International. 18.
8. Ibid., 24.
9. May-Chahal, C., & Palmer, C. (2018). Rapid Evidence Assessment: Characteristics and vulnerabilities of victims of online-facilitated child sexual
abuse and exploitation. Independent Inquiry into Child Sexual Abuse. UK: Lancaster University.
10. Stoilova, M., Livingstone, S., Khazbak, R. (2021). Investigating Risks and Opportunities for Children in a Digital World: A rapid review of the
evidence on children’s internet use and outcomes. Innocenti Discussion Papers no. 2021-01, Florence: UNICEF Office of Research – Innocenti.
12. The only two legally binding international instruments containing an obligation to criminalise the grooming of children for sexual purposes are:
Council of Europe. (2007). Convention on the Protection of Children against Sexual Exploitation and Sexual Abuse. Council of Europe Treaty Series
– No. 201. Article 23; and European Parliament and Council. (2011). Directive 2011/92/EU on combating the sexual abuse and sexual exploitation of
children and child pornography, and replacing Council Framework Decision 2004/68/JHA. Article 6.
ABOUT ONLINE CHILD SEXUAL EXPLOITATION AND ABUSE
increasingly common for offenders to sexually digital technologies, without necessarily having the
abuse children by, for example, manipulating them intention of meeting them and abusing them in
into self-generating and sharing CSAM through person.
Figure 2: Framing the main forms of online child sexual exploitation and abuse explored
by Disrupting Harm.
Internet or
communication
technology involved
Sexual exploitation Child sexual

and abuse abuse material
(physical contact)
Live-streaming
Grooming / coercion
The Disrupting Harm reports also address other using sexual content of the child that has previously
phenomena that contribute to understanding the been obtained as leverage.
contexts and socio-cultural environments in which
OCSEA occurs. Sexual harassment of a child15 and unwanted exposure
of a child to sexual content16 are other phenomena
The sharing of self-generated sexual content which can impact and enable OCSEA in some instances.
involving children13 can lead to or be part of OCSEA, For example, offenders can deliberately expose
even if this content is initially produced and shared children to sexual content as part of grooming to
voluntarily between peers, as it can be passed on desensitise them to sexual acts. However, for the
without permission or obtained through deception purposes of evidence-based policy and program
or coercion. development, it is important to acknowledge that
there are differences between voluntary viewing of
Sexual extortion of children14 refers to the use of sexual content by children and viewing that is forced or
blackmail or threats to extract sexual content or coerced. The former is not included in the definition of
other benefits (e.g., money) from the child, often OCSEA used in the Disrupting Harm study.
13. Cooper, K., Quayle, E., Jonsson, L. & Svedin, C.G. (2016). Adolescents and self-taken sexual images: A review of the literature. Computers in Human
Behavior, vol. 55, 706-716.
15. Ibid., 21.
16. Ibid., 44.
ABOUT KENYA – DEMOGRAPHICS AND INTERNET USAGE
Despite increasing connectivity around the world, few The available data presented here provide an important
countries regularly update their formal internet use statistics backdrop for understanding the various facets of
or disaggregate them for their child populations. This children’s internet use. However, methodological
presents a challenge in understanding how young people’s limitations affecting data quality for some secondary
lives are impacted by digital technologies, particularly in sources should be kept in mind. Relying on purposive
low- and middle-income countries. The infographic below or other non-probability sampling techniques means
summarises the latest data on internet access and social that the data cannot be considered representative of
media use in Kenya, some of which was gathered directly the population in question. In other cases, variations in
through the Disrupting Harm nationally representative data collection methods and definitions of internet use
household survey of internet-using 12-17-year-olds. pose a challenge for cross-country comparisons.
POPULATION TOTAL 2019 FEMALE POPULATION 2019 MALE POPULATION 2019

UN data: UN data: UN data:
52,574,000 26,452,000 26,122,000

17 20 23
(2018: 51,393,000) (2018: 25,859,000) (2018: 25,534,000)

18 21 24
Census data: Census data: Census data:
47,564,000 24,015,000 23,548,000

19 22 25
Under 18
POPULATION UNDER 18 2018 MEDIAN AGE 2020
27
20.1
UN data:
23,965,000
26
Census data:
47%
NO DATA
Urban
URBAN POPULATION
CHILDREN IN RURAL AREAS IN KENYA
2018: 27%
28
2030 prospect: 33.4%29

CONTINUE TO HAVE MUCH LESS ACCESS
TO HEALTH, SECURITY & EDUCATION THAN
27% THOSE LIVING IN URBAN AREAS.30
17. United Nations Population Division. (n.d.). World Population Prospects 2019. 24. Ibid.
18. Ibid. 25. Republic of Kenya. (2019). 2019 Kenya Population and Housing Census.
19. Republic of Kenya. (2019). 2019 Kenya Population and Housing Census. Volume I: Population by County and Subcounty.
Volume I: Population by County and Subcounty. 26. UNICEF. (2019). The State of the World’s Children 2019. UNICEF, New York.
20. United Nations Population Division. (n.d.). World Population Prospects 2019. 27. United Nations Population Division. (n.d.). World Population Prospects 2019.
21. Ibid. 28. United Nations Population Division. (n.d.). World Urbanization Prospects:
22. Republic of Kenya. (2019). 2019 Kenya Population and Housing Census. The 2018 Revision.
Volume I: Population by County and Subcounty. 29. United Nations Population Division. (n.d.). World Population Prospects 2019.
23. United Nations Population Division. (n.d.). World Population Prospects 2019. 30. UNICEF. (2018). Country Office Annual Report 2018 – Kenya.
Source: Disrupting Harm data
TOTAL INTERNET SUBSCRIPTIONS31 % via mobile INTERNET USE
subscriptions
AMONG CAREGIVERS
Oct –Dec 2020:
OF INTERNET-
44,391,490
32
USING CHILDREN
99%
For comparison, in Oct-Dec 2019
the number of subscriptions stood
51%
at 39,657,090 – also 99% mobile33
n = 1,014 caregivers of internet-using children.
MOBILE (SIM) PENETRATION

Dec 2020:
61.4 MILLION = 129.1%
34
Dec 2019:
54.5 MILLION = 114.8%
35
2020 INTERNET PENETRATION
RATES AMONG 12-17-YEAR-OLDS33
Total
67%
MOST POPULAR DEVICE TO ACCESS
†
12-13 Years THE INTERNET AMONG 12-17 Y.OS
55%
14-15 Years Mobile
99%
62%
13%
16-17 Years 6%
83%
Tablet Computer
Girls
66%
Boys n = 1,014 internet-using children.
Multiple choice question
†
68%
Rural Source: Disrupting Harm data
64%
MOST POPULAR PLACE TO ACCESS
Urban †
THE INTERNET AMONG 12-17 Y.OS
80%
n = 1,879 households. Home
GDP PER CAPITA 2019 (US$) 20% 27% 99%

$1,816.5
36
Internet Other
café
8% 15%
ONE OF THE FASTEST GROWING ECONOMIES Mail

School
IN SUB-SAHARAN AFRICA IN 201937
n = 1,014 internet-using children.
Multiple choice question
†
31. The number of subscriptions is not reflective of the number of unique users. 34. Communications Authority of Kenya. (2020). Second Quarter Sector
32. Communications Authority of Kenya. (2020). Second Quarter Sector Statistics Report for the Financial Year 2020/2021 (October-December 2020).
Statistics Report for the Financial Year 2020/2021 (October-December 2020). 35. Ibid.
33. Communications Authority of Kenya. (2019). Second Quarter Sector Statistics 36. World Bank. (2020). GDP per capita (current US$) – Kenya.
Report for the Financial Year 2019/2020 (October-December 2019). 37. World Bank. (2021). Kenya Overview.
POVERTY RATES Below poverty line Children in poverty
THE INEQUALITY GAP (HIGHEST

36.1% 45% RATES IN RURAL AREAS) IS ONE
OF THE HIGHEST IN AFRICA
2014 2015 2017

Approximately 8,300 36.1% of the population According to the UNICEF Kenya 2017 Annual Report,
people owned 62% of still lived below the 45% of children under 18 (9.5 million children) were
the country’s wealth).38 national poverty line39 experiencing poverty in Kenya (stark inequities – 85%
in Turkana county compared to 7% in Nairobi)40
HUMAN DEVELOPMENT INDEX The Human Development 41
Index (HDI) is a summary OFFICIAL LANGUAGES

measure of average achievement
2019 score: in key dimensions of human
ENGLISH
development: a long and healthy
life, being knowledgeable and
having a decent standard of
SWAHILI
0.601
living. The HDI is the geometric
rank: mean of normalised indices for
each of the three dimensions.
143/188
FREQUENCY OF INTERNET USE AMONG 12-17 YEAR OLDS
40% Less than once a month

At least monthly
30% At least weekly
Once a day or more
20%
10%
0%
Total 12-13 14-15 16-17 Boy Girl Urban42 Rural
38. Beegle, K., Christiaensen, L., Dabalen, A., and Gaddis, I. (2016). Poverty in a Rising Africa.
39. World Bank. (n.d.). Poverty & Equity Data Portal.
40. UNICEF. (2017). Country Office Annual Report 2017 – Kenya.
41. Republic of Kenya. (2010). The Constitution of Kenya, Article 7.
42. The urban category includes the urban and peri-urban areas of the sample.
FREQUENCY OF INTERNET USE AMONG

MOST POPULAR SOCIAL MEDIA PLATFORMS
43
CAREGIVERS OF INTERNET-USING CHILDREN

WhatsApp
88.6%
Facebook
At least once a day
28% 88.5%
YouTube
At least weekly 9%
48% Never 51.2%
Google+
At least monthly: 3% 41.3%
12%
Less than once a month Instagram
39%
Twitter:
27.9%
n = 1,014 caregivers of internet-using children.
Yahoo
18.6%
LinkedIn
CHILDREN WHO USE SOCIAL MEDIA
9.3%
ON A WEEKLY BASIS Snapchat
9.0%
51% 33% 52% 63% 55% 47%
n = 3,269 Kenyans aged 14-55
Total
16-17
14-15
Boys
12-13
Girls
ICT DEVELOPMENT GLOBAL

INDEX RANKING (ITU)
44
n = 1,014 internet-using children. CYBERSECURITY
INDEX RANKING 2018
45

2/38
CHILDREN WHO USE INSTANT
MESSAGING APPS ON A WEEKLY BASIS 44/175*
13/38
39% 24% 36% 54% 45% 35%
General ranking
In the world
Total
16-17
14-15
Boys
12-13
Girls
138/175
Africa
Africa

* behind Mauritius46
MARKET SHARES IN MOBILE DATA

47
SUBSCRIPTIONS (DEC 2020) 26.5% Safaricom
5.2% PLC
0.4%
Jamii Telcom Airtel
67.6%
0.3% Telecommunications Kenya Networks
Equitel Ltd Limited Limited
43. United States International University – Africa. (2019). Social Media instruments and the level of technical and organisational measures taken to
Consumption in Kenya: Trends and Practices. reinforce international cooperation and cybersecurity.
44. International Telecommunication Union. (2017). ICT Development 46. International Telecommunication Union. (2019). Global Cybersecurity
Index 2017. Index (GCI) 2018.
45. The Global Cybersecurity Index measures the commitment of 47. Communications Authority of Kenya. (2020). Second Quarter Sector
countries to cybersecurity based on the implementation of legal Statistics Report for the Financial Year 2020/2021 (October-December 2020).
Overview of legislation and policy number of provisions to ensure that child victims
receive support and protection immediately after
The most relevant pieces of legislation currently the abuse is reported as well as during the legal
in effect regarding sexual offences, including proceedings.
OCSEA-related crimes, in Kenya are the Computer
Misuse and Cybercrimes Act48 and the Sexual Kenya has two national policies already in
Offences Act49. effect which touch on OCSEA: The National
Plan of Action Against Sexual Exploitation of
The Computer Misuse and Cybercrimes Act50 Children in Kenya, 2018-2022 and the National
provides a quite comprehensive definition of Information, Communications and Technology
CSAM and explicitly criminalises acts associated Policy (2019). The former includes an objective
with it51 as well as the attempt to commit these and activities related to the prevention of
crimes.52 The Sexual Offences Act sets the age of OCSEA, while the latter sets out the broad
sexual consent at 1853 but provides no close-in-age activities to be undertaken by the government
exemption for consensual sexual relationships on child online protection. However, interviews
between peers under 18. It also prohibits certain with representatives of the National Council of
forms of conduct associated with CSAM.54 Children’s Services (RA1-KY-02-A) and the Ministry
of Information Communication and Technology,
The provisions of these laws relating to CSAM Innovation and Youth Affairs (RA1-KY-12-A) made
could potentially be applied to live-streaming of clear that neither policy has been launched or
child sexual abuse. However, this is not explicitly disseminated widely to stakeholders.
stated and there is no separate provision
criminalising live-streaming of abuse. Similarly, Two policies exclusively concerned with child
there are no provisions prohibiting online online protection are under development –
grooming for sexual purposes and sexual extortion namely, the National Plan of Action on Online
committed in the online environment. Child Sexual Exploitation and Abuse and the
National Strategy on Child Online Protection.
The upcoming Children Bill 2021 was submitted The former is spearheaded by the Department
to Parliament in 2020 but is yet to be enacted. of Children’s Services and is in the final stages of
According to interviews with three representatives development.56 It is anchored in the WeProtect
from the Kenya Law Reform Commission, it will Model National Response. The National Strategy
comprehensively criminalise online grooming and on Child Online Protection, which is led by the
may also be applicable to live-streaming of child Communications Authority of Kenya, will embody
sexual abuse. the ITU Guidelines on Child Online Protection and
will be formulated upon cabinet approval.
The Victim Protection Act55 aims to secure victims
of crimes from further harm. The Act includes a
48. Republic of Kenya. (2018). The Computer Misuse and Cybercrimes Act No. 5 of 2018.
49. Republic of Kenya. (2006). The Sexual Offences Act No. 3 of 2006. (Last revised in 2019).
50. Republic of Kenya. (2018). The Computer Misuse and Cybercrimes Act No. 5 of 2018. Section 24 (3).
51. Ibid., Section 24.
52. Ibid., Section 42(2).
53. Republic of Kenya. (2006). The Sexual Offences Act No. 3 of 2006. Section 8. (Last revised in 2019).
54. Ibid., Section 16. (Last revised in 2019).
55. Republic of Kenya. (2014). Victim Protection Act No. 17 of 2014.
56. A representative from the Department of Children’s Services confirmed that all the planned stakeholder consultations and focus group
discussions had been completed and a draft policy developed. The only remaining activity was a validation, which was tentatively planned for May
of 2021.
1. CHILDREN
ONLINE IN KENYA
The main focus of this report is to present the perspectives of
young people and duty-bearers around the sexual exploitation
and abuse of children facilitated or committed through
digital technologies. However, it is important to situate these
offences within the wider context of children’s internet use in
Kenya. This first chapter therefore, presents a brief overview
of children’s internet access and the activities enjoyed by the
majority of children online before going on to describe the
occurrence of riskier online activities and the ways in which
these are perceived by children and their caregivers.
1.1 INTERNET ACCESS AND BARRIERS
Sampling data from the Disrupting Harm household Among internet-using children, 60% go online at
survey suggest that 67% of 12–17-year-olds in Kenya least once a week. As is the pattern in other countries
are internet users – i.e. they have used the internet around the world,59 older children are more frequent
within the past three months.57,58 This figure rises users. Boys go online somewhat more frequently
from 55% among children aged 12-13 and 62% than girls (see Figure 3). Children living in urban areas
among children aged 14-15 to 83% among children use the internet more frequently than children in
aged 16-17. Boys and girls are just as likely to be rural areas.
internet users. In rural areas, 64% of children are
internet users compared to 80% in urban areas.
Figure 3: Frequency of children’s internet use
Total 27% 13% 33% 27%
12–13 32% 15% 33% 20%
14–15 25% 13% 39% 23%
16–17 26% 11% 28% 35%
Boy 22% 13% 35% 30%
Girl 32% 13% 30% 24%
0% 20% 40% 60% 80% 100%
Less than once a month At least monthly At least weekly Once a day or more
Base: Internet-using children aged 12-17 in Kenya. n = 1,014.
57. While conducting the random walk to identify eligible children to partake in the main survey, we also collected data from every household
visited about the number of 12–17-year-old children living there, their gender, age, and whether they had used the internet in the past three
months. This allowed us to estimate internet penetration rates for all 12–17-year-old children in Kenya. n = 1,879 households.
58. The question used to determine whether a 12-17-year-old was an internet user: Has [PERSON] used the internet in the last three months? This
could include using a mobile phone, tablet or computer to send or receive messages, use apps like Facebook, WhatsApp, Instagram, send emails,
browse, chat with friends and family, upload or download files, or anything else that you usually do on the internet.
59. See: Global Kids Online: http://globalkidsonline.net/.
1.1 INTERNET ACCESS AND BARRIERS
Almost half of the caregivers surveyed have never limited online experience, it is important to consider
used the internet. Those aged 50 and above are the support and knowledge they need, as well as the
far less likely to be internet users than younger role that can be played by schools in guiding their
caregivers. Men are rather more frequent users than children’s use of the internet.
women (see Figure 4). As many caregivers have
Figure 4: Frequency of caregivers’ internet use
Total 48% 12% 3% 9% 28%
< 29 22% 13% 4% 14% 47%
30-39 48% 13% 3% 8% 28%
40-49 52% 9% 2% 9% 27%
50+ 73% 15% 3% 3% 6%
Men 44% 13% 2% 7% 33%
Women 50% 11% 3% 10% 25%
0% 20% 40% 60% 80% 100%
Never Less than once a month At least monthly At least weekly Once a day or more
Base: Caregivers of internet-using children aged 12-17 in Kenya. n = 1,014.
Most children use the internet from home, which is About four out of every five children who use a
consistent with data from other countries. Only 15% smartphone share it with someone else. Among
of children have ever used the internet at school, and those children who use computers to go online,
very few do so regularly. almost all of them (93%) share the computer with
someone else. Only 16% of girls have their own,
As in most other countries, smartphones are by far unshared smartphone compared to 28% of boys. In
the most common device used by 12–17-year-old rural areas, 19% of internet-using children have their
internet users to go online, probably due to their own smartphone compared to 27% in urban areas.
relatively low cost and portability.60 As many as 99%
use smartphones, while 13% also use computers and Almost all internet-using children face barriers in
6% tablets. There are no notable differences by age, accessing the internet and only 5% have readily
gender or urban-rural location.
60. Livingstone, S., Kardefelt Winther, D., & Saeed, M. (2019). Global Kids Online Comparative Report. Innocenti Research Report. Florence: UNICEF
Office of Research – Innocenti.
available access whenever they want or need it. High go online when they want or need to because
internet and data costs are barriers to access for 50% someone else is using the digital device (see
of internet-using children, while 45% are unable to Figure 5).
Figure 5: Barriers to access for internet-using children.
70%
60%
50%
50% 45%
40%
30% 22% 21% 19%
20%
10% 7% 5% 3%
0%
Paying for The device Parents Limited Slow Teachers Always Other
internet/ they use to don’t electricity connection don’t allow have barriers
data is too go online is allow where or poor them access
expensive being used them they live signal where
by some- they live
one else
Internet access is generally a little easier for older Overall, girls are more likely than boys to name
children with the main obstacle for 16-17-year-olds parental restrictions (25% girls; 18% boys) and
being high data costs. This may reflect the fact that shared devices (50% girls; 40% boys) as barriers to
older children use the internet more frequently internet access.
than younger children and engage in more activities
online (see Figure 6), therefore requiring more data.
1.2 CHILDREN’S ACTIVITIES ONLINE
The most popular online activities among the children surveyed are watching
videos, using social media, instant messaging, online gaming and watching live-
streams, followed by going online for school work and to look up new information.
Older children generally engage in a wider range of online activities than younger
children. However, gaming is most popular among the youngest children.
It is worth considering that these categories are not minor, as has been observed in other countries.61
intended to be mutually exclusive – for example, a As an exception, 40% of boys played online games
child could go online to watch a video as part of their compared to only 28% of girls. Girls were also less
school work. Nonetheless, Figure 6 below provides likely than boys to use instant messaging and social
a greater understanding of how 12-17-year-olds in media while boys were more likely to search for news
Kenya use the internet and the activities they enjoy. online compared to girls (see Figure 6).
Gender differences in online activities are relatively
Figure 6: Activities children engage in online at least once a week.
Online activities Total 12-13 14-15 16-17 Boy Girl
Watching videos 57% 56% 61% 53% 58% 55%

Using social media 51% 33% 52% 63% 55% 47%
Using instant messaging 39% 24% 36% 54% 45% 35%
Playing online games 34% 44% 33% 27% 40% 28%
Watching a live-stream 34% 38% 34% 30% 36% 31%
School work 32% 28% 31% 37% 32% 32%
Searching for new information 25% 16% 26% 31% 30% 21%
Following celebrities and public figures
20% 11% 20% 26% 21% 19%
on social media
Searching for news 19% 11% 21% 24% 24% 15%
Talking to family or friends who live
17% 13% 16% 21% 18% 16%
further away
Participating in a site where people share
15% 9% 17% 18% 17% 14%
their interests
Searching for information about work
13% 8% 11% 20% 13% 14%
or study opportunities
Creating their own video or music 11% 12% 11% 10% 10% 11%
Searching for health information 10% 8% 9% 12% 10% 10%
Seeking emotional support 7% 4% 6% 9% 6% 8%
Looking for information on local events 6% 4% 8% 7% 8% 5%
Discussing political or social problems 6% 3% 6% 8% 6% 5%
Creating a blog or website 3% 1% 3% 5% 3% 4%
61. Ibid.
1.3 PERCEPTIONS AND EXPERIENCES
OF RISKY ONLINE ACTIVITIES
Discussion of online risks for children often hinges upon adult-centric

perceptions. To ensure we also understood children’s perceptions, we asked
them and their caregivers about their engagement in, and perceptions of,
various online risky activities.
1.3.1 Contact with strangers online simply be how young people now make new
and in person friends. Alternatively, this could indicate a lack of
Of the caregivers in our household survey, 82% rated awareness of how speaking to strangers online
talking to online strangers as ‘very risky’ for children, might lead to harmful outcomes.
but children themselves were less concerned. Just Child-centred workshops conducted in relation
44% of internet-using children considered this to the Disrupting Harm project with 27 children
activity ‘very risky’ for children of their age. Children in Kenya suggested that children’s understanding
aged 12-13, and girls, were most likely to describe of online ‘strangers’ can be rather nuanced. While
talking to someone on the internet who they have aware of the need to be careful, they reflected
never met in person, as ‘very risky’ (see Figure 7). that a stranger has the potential to be good or
Similarly, 88% of the caregivers surveyed thought bad. Many children said that whether or not they
it ‘very risky’ for children to send their personal felt safe interacting with a stranger depends on
information to someone they have never met face-to- the context of their interactions.62
face, compared to 57% of the children. Turning to the actual behaviour of the children in
While most children recognised that interacting our survey in relation to people they first met online,
with strangers carries some level of risk, a substantial 35% had added people they had never met before
proportion said that these activities were ‘not risky to their contact lists. Over one in four internet-using
at all’ or were unsure about it. This may be because children had shared their personal information with
many such connections are not harmful and may someone they had never met face-to-face.
Figure 7: Children's risk assessment of speaking to online stranger versus children who
have added strangers to their friends list in the past year.
Talking to someone on the internet whom they I added people who I have never met
have not met face-to-face before face-to-face to my friends or contacts list
% of children who say % of children who have

44% this is 'very risky' for
children their age 35% done this in the past year
62. Third, A., Moody, L., & Theakstone, G. (2020). Children‘s Digital Experiences: Kenya Country Report.
1.3 PERCEPTIONS AND EXPERIENCES OF RISKY ONLINE ACTIVITIES
Figure 8: Children's risk assessment of sharing their personal information with online strangers
versus children who have engaged in this behaviour in the past year.
Sending personal information (e.g., their full I sent my personal information (e.g., my full
name, address or phone number) to someone name, address or phone number) to someone
they have never met face-to-face I have never met face-to-face
% of children who have

% of children who say done this in the past year
children their age 22%
In our household survey, we also queried perceptions this behaviour as ‘not risky at all’.
about situations where getting to know people
online leads to face-to-face encounters. Over half of Within the past year, 14% of the children surveyed
the children and as many as 86% of their caregivers had met someone in person whom they had first
thought that meeting ‘online strangers’ in person is met online. Out of these children, the great majority
‘very risky’ for children. Girls were more likely than were happy about the experience (see Figure 10).
boys to regard this as ‘very risky’ behaviour (60% vs. Research done across more than 30 countries
45%, respectively). However, 16% of children viewed around the world has produced similar findings.63,64
Figure 9: Children's risk assessment of meeting online strangers in person versus children who
have engaged in this behaviour in the past year
Going to meet someone face-to-face that In the past year, have you ever met anyone
they first got to know online face-to-face that you first got to know on the
internet?

done this in the past year
% of children who say

63. Ibid.
64. Smahel, D., Machackova, H., et al. (2020). EU Kids Online 2020: Survey results from 19 countries. Florence: UNICEF Office of Research – Innocenti.
Figure 10: How children felt the last time they met someone face-to-face whom they had first got
to know on the internet.
70%
61%
60%
50%
40%
40%
30%
20%
9%
6% 6% 7% 7%
10%
2%
0%
Happy Excited Proud Thoughtful Sad Angry Fearful Anxious
Base: Children who, within the past year, have met someone face-to-face whom they first got to know on the internet. n = 139.
There are clearly incongruences between children’s to group events with caregivers. The experiences of
and caregivers’ perceptions. Clearly, meeting most internet-using children in Kenya and other
someone you do not know face-to-face for the first countries around the world seem to indicate that the
time can be very risky. This report mentions some risk of harm is relatively low for children in general,
cases which had severe consequences for doing so. although the harm might be severe if it occurs. While
Such cases probably explain why caregivers are so many children in Kenya are aware that engaging
worried. But there are many different types of such with online strangers carries a level of risk, we need
encounters, like connecting with new children in the to ensure all children are informed and taught how
community first online and then in person, or going to engage safely and responsibly.
Is Restricting Children's Internet Access the environment in the long term. On the other
Answer? hand, supportive engagement by adults has been
Many caregivers instinctively react to online risks by associated with positive skills development for
restricting their children’s internet use in a bid to children in other countries.65 Supportive mediation
protect them. Such restrictive practices seem quite could include engaging in activities together, talking
common in Kenya. For example, 38% of the children to children about their internet use, and educating
in our survey reported that they are not allowed them about the risks that exist online and how best
to use social media, and 20% are not allowed to to avoid them. In these ways, we allow children to
watch videos online. In addition, 23% said that benefit from the many useful activities and skills that
their caregivers often limit how long they can stay the internet has to offer, while providing parental
online. When asked what they would do if their guidance and support. While caregivers in Kenya use
child was bothered by something online, one third the internet less frequently than their children and
of caregivers said they would restrict their child’s may worry that they do not have enough knowledge
internet access. to guide them, they can still talk to their children
about what they do online and provide an open and
This approach might reduce children’s exposure supportive home environment. Information about
to online risks in the short term, but it also reduces online risks and how to avoid them might also be
their digital skills and familiarity with the online provided by schools or specialised organisations.
65. Livingstone, S., Kardefelt Winther, D., & Saeed, M. (2019). Global Kids Online Comparative Report. Innocenti Research Report. Florence: UNICEF
Office of Research – Innocenti.
1.3.2. Seeing sexual images online over their child becoming a victim of a crime and
As shown in Figure 11, when the caregivers surveyed whether they would be able to provide for their
were asked to select their top three concerns for child (see Figure 11). Among the children surveyed, a
the child being interviewed, seeing sexual images substantial 63% believed that seeing sexual images
was the most common concern, along with their or videos on the internet is ‘very risky’ for children
child’s health. Worries over sexual content, concerns their age, but among caregivers, the ratio was 89%.
Figure 11: Caregivers’ top concerns regarding their children.
70%
60% 54% 55%
50% 44%
39%
40% 34%
29%
30%
20% 12%
10%
0%
My Having Other children My child A stranger My child My child
child’s enough treating my seeing sexual contacting becoming a revealing
health money to child in a images or my child on victim of personal
care for my hurtful or videos on the the internet crime information
child nasty way internet online
The frontline workers whom we surveyed regarded OCSEA, ahead of issues like migration, experiences
‘access and exposure to pornography’ as the most of family and community violence, or living on the
important factor increasing children’s vulnerability to street (see Figure 12).
Figure 12: Frontline workers’ perceptions of factors affecting children’s vulnerability to OCSEA.
Access and Exposure to Pornography

2%
98%
Dropping out of school
4%
96%
Increased access to technology and Internet
6%
94%
Extreme poverty
8%
92%
The child themselves having to migrate for work
10%
90%
Being left behind by parent/guardian who has migrated for work
12%
88%
Gender Norms
16%
84%
Family violence
18%
82%
Cultural practices
20%
80%
Living and/or working on the street
22%
78%
Living with one or multiple disabilities
24%
76%
Community Violence
30%
70%
Belonging to an ethnic minority group
54%
46%
0% 20% 40% 60% 80% 100%
Disagree Agree
Base: Frontline workers. n = 50.
Accidental or intentional glimpses of sexual content In practice, 37% of internet-using children in Kenya
are one thing; being exposed to sexual images as who took part in the household survey said they had
part of a grooming process intended to desensitise seen sexual images or videos at least once in the
the child and pave the way for subsequent requests past year. Only 16% reported actively looking for such
for images or sexual acts is another. While viewing material online, while 33% were exposed to sexual
violent or degrading sexual content can serve images or videos when they did not expect sexual
to normalise harmful gender norms and sexual content online.
behaviour, seeing some pornography appears to
be an increasingly present experience for young
people.66 Addressing both phenomena is needed.
Figure 13: Children's risk assessment of seeing sexual images or videos online versus children who
have actively looked for this content in the past year
Seeing sexual images or videos on I have seen sexual images or videos online
the internet because I wanted to (for example, I accessed a
website or social network expecting to find that
kind of content there)

% of children who say done this in the past year
children their age
16%
Among the children who reported seeing sexual videos on their social media feeds, while one in five
images or videos online by accident, 42% said they saw saw them in online advertisements. Around 22% of
these images or videos on their social media feeds. children said the images or videos were sent to them
Around one in five saw them in online advertisements directly.
and 22% of children said the images or videos were
sent to them via direct messaging apps. 1.3.3 Making and sharing self-generated sexual
content
Older children were more likely than younger children Most of the children and caregivers surveyed agreed
to see sexual images or videos on social media when with the statement “it is wrong for a person to take
they did not expect it. There were no major differences naked images or videos of themselves”. In addition,
in accidental exposure to sexual content on social 91% of caregivers and 68% of children thought it was
media by gender. Among the 21% of children who ‘very risky’ to share a sexual image or video online,
reported seeing sexual images or videos online by while only 3% and 12%, respectively, thought it was
accident, nearly half said they saw these images or ‘not risky at all’.
66. See for example: Crabbe, M. & Flood, M. (2021). School based Education to Address Pornography’s Influence in Young People: A Proposed
practice framework. American Journal of Sexuality Education 16(1).
In practice, 6% of the children in the household of themselves in the past year. There were no clear
survey said they had shared naked pictures or videos differences by age or gender.
Figure 15: Children's risk assessment of sending sexual content online versus children who have
engaged in this behavior in the past year
Sending a sexual image or video to someone In the past year, how often have you shared
on the internet naked pictures or videos of yourself with
someone else online?

done this in the past year
% of children who say

The main reasons given by children for sharing naked Seven of the 60 children had shared self-generated
pictures or videos were that they were in love, flirting sexual content because they were threatened and
or having fun, that they trusted the other person, and six because they were being pressured by their
that they found nothing wrong with sharing such friends (see Figure 16). Figures from the survey are
content. Five percent had allowed someone else to representative of 12-17-year-old internet users. When
take naked pictures or videos of them. scaled up to this population of children, the numbers
are far greater.
Figure 16: Reasons given by children for sharing naked images or videos of themselves.
In love
30%
Flirting or having fun

22%
Trusted the other person

15%
Did not think there was anything wrong with sharing the pictures or videos
13%
Worried that I would lose the person if I didn't share

12%
Threatened
12%
Pressured to share the pictures or videos by their friends

10%
Offered money or gifts in exchange for the pictures or videos

7%
Prefer not to say

7%
Don't know
5%
Wanted the attention of the person

2%
0% 5% 10% 15% 20% 25% 30%
Base: Children who have shared naked images or videos of themselves in the past year. n = 60.
Most of the 60 children had shared the images or survey have not received any information on how to
videos with a friend or someone else they knew in stay safe online. To ensure not only that children are
person (35%), or with a romantic partner (30%), but aware of possible risks but that they know what to
some children (12%) had shared them with someone do about them, there is a need for comprehensive
they met online who had no other connection with digital literacy and safety training. This should include
their life. information about what children can do if they are
being bothered online, what content to share and
Overall, although children in Kenya show some level not to share with others, and basic skills such as how
of awareness about online risks, 56% of the internet- to change their privacy settings and block people
using children who took part in the household from contacting them.
The rise in self-generated sexual content Figure 14: Mapping the consequences of
involving children sharing self-generated sexual content
The increasing use of technology is leading to involving children.
shifts in notions of privacy and sexuality among
children in some parts of the world, particularly
among adolescents as they mature.67 Behaviours
that are increasingly normative to young people
can be bewildering for adults who grew up in a Self-
different time. For example, chatting and video generated
sexual content
live-streaming is frequent, whether among small
private groups of friends or large, anonymous Live-streaming
Pictures
public audiences. While much of this is harmless,
Videos
producing and sharing self-generated sexual
content using these tools is also increasing, and
bringing significant risks.68
The sharing of self-generated sexual content

by children is complex and includes a range of
different experiences, risks and harms. As our data Non-
Coerced
show, some self-generated content is created and coerced
shared by adolescents voluntarily. Such exchanges
are increasingly becoming part of young people’s
sexual experiences. However, our data also shows
that the creation and sharing of self-generated
sexual content can be coerced, for example through
grooming, threats or peer-pressure (see chapter 2.2).
While coercion can clearly be seen as a crime

and leads directly to harm, there can be negative NO HARM HARM
consequences for children sharing any sexual
content including in cases where sharing is not
coerced. Material shared voluntarily may not
cause harm at first, but there remain risks if it is
later shared beyond the control of the person
who created it. Once it exists, such content can
also be obtained deceptively or using coercion
and circulated by offenders perpetually.69,70
(see Figure 14).
67. Livingstone, S. & Mason, J. (2015). Sexual Rights and Sexual Risks among Youth Online: A review of existing knowledge regarding children and
young people’s developing sexuality in relation to new media environments. London: European NGO Alliance for Child Safety Online.
68. Thorn & Benson Strategy Group. (2020). Self-Generated Child Sexual Abuse Material: Attitudes and experiences. U.S.: Thorn.
69. Bracket Foundation. (2019). Artificial Intelligence: Combating Online Sexual Abuse of Children. 10.
70. EUROPOL. (2019). Internet Organized Crime Threat Assessment 2019. Netherlands: EUROPOL.
2. ONLINE CHILD
SEXUAL EXPLOITATION
AND ABUSE IN KENYA
Following on from children’s perceptions of, and participation in,
various risky online practices, this chapter will turn to the threat
of online child sexual exploitation and abuse (OCSEA) in Kenya.
We draw on a variety of sources – including law enforcement data,
mandated reports from U.S.-based technology companies to
the NCMEC related to Kenya, surveys with frontline workers and
surveys, interviews and conversations with children themselves
– in order to create a well-rounded presentation of the nature of
these crimes against children.
2. ONLINE CHILD SEXUAL EXPLOITATION AND ABUSE IN KENYA
This chapter presents estimates of the occurrence of certain instances of OCSEA

based on data from law enforcement units (chapter 2.1) and children’s self-reported
experiences (chapter 2.2 and 2.3). For several reasons, estimates are not intended
to provide a conclusive picture of the prevalence of OCSEA. Firstly, the existing
administrative data that we have accessed, such as that kept by law enforcement
authorities, rarely delineates or classifies OCSEA elements. Secondly, with respect
to the household survey, we would expect a degree of under-reporting due to
privacy concerns and the discomfort of openly discussing sex. Furthermore, in
households where sexual abuse occurs, we expect we would be less likely to be
provided access to survey children. Finally, many estimates are based on analysis
of sub-samples of the survey data which are small because OCSEA is still a rarely
reported phenomenon, which results in a larger margin of error.
While we have full confidence in our data and the reader interprets the findings in this chapter as a
quality of the sample obtained, the challenges good approximation of the occurrence of certain
of researching specific and sensitive phenomena crimes against children related to OCSEA in Kenya
means that we inevitably lose precision in the final and the extent to which internet-using 12–17-year-old
estimate. For these reasons, we suggest that the children in Kenya are subjected to OCSEA.
2.1 LAW ENFORCEMENT DATA
The analysis in this chapter draws on qualitative and quantitative data from law
enforcement authorities and a number of partner organisations, with a view to
understanding offences relevant to OCSEA that were recorded in the country,
offender and victim behaviours, crime enablers and vulnerabilities.
2.1.1 Recorded OCSEA offences AHTCPU. It appears that the discrepancy stems from
The Anti Human Trafficking and Child Protection AHTCPU only recording cases that they consider
Unit (AHTCPU) of the Kenyan National Police Service actionable.
reported the following case numbers for 2017, 2018 It is likely that the difference lies in the number
and 2019: of so-called ‘meme’ reports: CyberTips based on
material circulated on social media in bad taste,
Figure 17: Number of CSEA/OCSEA cases inappropriate humour, or in a misguided attempt to
recorded by AHTCPU help the child. While the material or activity may not
be actionable by law enforcement, it does constitute
2017 2018 2019 a breach of the terms of service of the reporting
Number of offline 25 33 21 electronic service provider, and is thus reported.
CSEA cases
2.1.2 International OCSEA detections and
Number of OCSEA - 3,160 4,133 referrals
cases
On behalf of Kenyan law enforcement, data was
Base: Data provided by AHTCPU. requested for Disrupting Harm from NCMEC on
CyberTips concerning suspected child sexual
Because offline instances of child sexual exploitation
exploitation in Kenya.
and abuse (CSEA) are usually reported to local police
stations, the number of CSEA cases investigated U.S. federal law requires that ‘electronic service
by the national specialist unit (presented in Fig providers’ (i.e., technology companies) based in the
16 above) is not representative of the country as a U.S. report instances of suspected child exploitation
whole. The larger numbers of OCSEA cases may be to NCMEC’s CyberTipline. However, for providers not
explained by the fact that the national specialist based in the U.S., this reporting is voluntary and not
unit is the recipient of reports made via NCMEC (see all platforms report suspected child exploitation
Fig 17 below). The Unit did not begin to investigate to NCMEC. There is therefore a data gap for several
OCSEA until 2018, hence the recording of 0 cases in platforms that are popular in the Disrupting Harm
the previous year. focus countries. Furthermore, it must be considered
that this CyberTip data only represents cases
The actual number of NCMEC CyberTipline reports
reported to NCMEC, and not a full picture of the
(CyberTips) recorded by AHTCPU were 3,160 and
extent of OCSEA in Kenya. CyberTipline reports under
4,133 in 2018 and 2019 respectively, much lower
this category may reference more than one file of
than the total CyberTips sent by NCMEC (16,108
CSAM. For example, some reporting ESPs include
and 12,788 in 2018 and 2019 respectively). There is a
more files per report, as opposed to one image per
discrepancy between the number of CyberTips sent
report and multiple reports per suspect.
by NCMEC and the number of recorded cases by
Figure 18: CyberTips concerning suspected child sexual exploitation in Kenya.
% Change % Change % Change

2017 2018 2019
2017- 2018 2018-2019 2017-2019
Kenya 12,361 16,108 12,788 30% -21% 3%
Global Total 10,214,753 18,462,424 16,987,361 81% -8% 66%
Kenya % of Global Total 0.12% 0.09% 0.08%
Base: Data provided by NCMEC.
Kenya shows a consistently low proportion of Kenya of 21% between 2018 and 2019 was more
suspected child sexual exploitation in CyberTips, marked than that for the global total (8%). This
an average of 0.09% in the years 2017-2019. This is may be indicative of a move in Kenya away from
lower than might be expected, given that Kenya misuse of the platforms that report suspected child
accounted for 0.68% of the world’s population, exploitation to NCMEC, thereby raising the further
and 1.00% of the world’s internet using population question of where OCSEA offenders might move to.
according to United Nations and International
Telecommunications Union estimates.71 Analysis of the types of incidents captured by
CyberTips reveals that the possession, manufacture
Kenya also saw a much smaller percentage increase and distribution of CSAM (referred to in U.S.
(just 3%) in CyberTips between 2017 and 2019 than legislation as ‘child pornography’) accounts for
the global distribution. Specifically, a reduction for almost all of the CyberTips for Kenya in the reporting
period (see Figure 19).
Figure 19: CyberTips concerning suspected child sexual exploitation in Kenya, by incident type.
Incident Type 2017 2018 2019
CSAM, including possession, manufacture and distribution 12,359 16,101 12,779

(NCMEC classification: child pornography)72,73
Travelling child sex offences 1 2 7

(NCMEC classification: child sex tourism)74
Child sex trafficking - - -
Child sexual molestation 1 2 2
Online enticement of children for sexual acts - 3 -
CyberTips classified as relating to CSAM increased in two reports for Kenya as ‘Priority 1’, indicating a child
2018 and declined in 2019 in line with the trend for in imminent danger.
Kenya’s total. In terms of priority level, NCMEC tagged
71. International Telecommunications Union. (n.d.). Statistics.

72. The terminology used by NCMEC is ‘child pornography’, to align with U.S. legislation. Disrupting Harm advocates use of the term ‘child sexual
abuse material’ in line with the Luxembourg Guidelines.
73. CyberTips under this category may reference more than one file of CSAM. For example, some reporting electronic service providers include
more files per report, as opposed to one image per report and multiple reports per suspect.
74. The terminology used by NCMEC is ‘Child Sex Tourism’, to align with U.S. legislation. Disrupting Harm advocates use of the term ‘Travelling Child
Sex Offences’ in line with the Luxembourg Guidelines.
Nearly 100% of NCMEC CyberTips for Kenya in reporting period. This would indicate some diversity
the period 2017 to 2019 had an electronic service in the usage of platforms by the general population,
provider as their source. A total of 29 electronic in line with the level of internet connectivity in the
service providers submitted at least one CyberTip country, and in the misuse of a range of platforms by
of suspected child exploitation for Kenya in the OCSEA offenders.
Figure 20: NCMEC CyberTips concerning suspected child sexual exploitation in Kenya, by
reporting electronic service providers.
Reporting Electronic Service Provider 2017 2018 2019 % of 2019 Total
Facebook 11547 15140 11592 90.66%
Instagram Inc. 397 570 770 6.02%
Google 366 332 349 2.73%
Tagged.com 19 12 27 0.21%
Snapchat 1 1 11
WhatsApp Inc. 1 4 11
Twitter Inc. / Vine.co 5 9 6
Pinterest Inc. 4 6 4
Microsoft – Online Operations 1 4 3
MediaFire 1
Multi Media, LLC/Zmedianow, LLC/Chaturbate 3 1
Hacker Factor 1
motherless 1
4chan community support LLC 1
Base: NCMEC CyberTips sorted by 2019 counts, null results removed.
Figure 20 shows that while Facebook accounts for CyberTips for Kenya from Chaturbate, a platform
93% of total CyberTips for Kenya in the reporting specialising in the provision of adult live-streamed
period as a whole, this proportion reduced slightly in sexual activity that is often paid for in tokens, raises
2019. This correlates with an increase in the number the possibility of OCSEA with a commercial element.
of CyberTips submitted by Instagram, which almost The presence in Kenya of OCSEA offenders with
doubled between 2017 and 2019. Although CyberTips a level of technical sophistication and specialist
from Facebook increased by 31% between 2017 and interest is demonstrated by the appearance in the
2018, they subsequently fell by 23%. Small increases data of self-avowed ‘moral free file host’ motherless,
in CyberTips from popular services such as Snapchat anonymous image-based bulletin board 4chan,
and WhatsApp were also observed. file sharing service Mediafire and digital forensics
research company Hacker Factor (1 CyberTip each).
The variety of platforms among the reporting
electronic service providers may also speak to the Data supplied by Kenyan law enforcement about
nature of suspected OCSEA offending. Multiple platforms used to commit recorded OCSEA offences
CyberTips from Tagged.com in 2017-2019 speak allows for comparison with those electronic service
to the persistent misuse of adult dating sites for providers reporting to NCMEC (see Figure 21).
suspected distribution of CSAM in Kenya. Four
Figure 21: Online services misused in OCSEA This distribution is broadly consistent with the
cases recorded by AHTCPU. NCMEC CyberTip data insofar as Facebook is
mentioned in 90% of cases. There are also notable
Service Misused differences, however, particularly in relation to
2018 2019
(mentions) WhatsApp. For this platform, the total number
Facebook 2,500 4,009 of mentions in OCSEA cases is larger than the
number of CyberTips it made in 2018 and 2019.
Instagram 110 400 Whereas CyberTips have a single source, a case
WhatsApp 11 60 recorded by law enforcement can involve the
misuse of more than one platform. Nevertheless,
Google 20 126
WhatsApp would appear to be more prominent in
Snapchat 0 2 the OCSEA caseload of Kenyan law enforcement
than is evident in the NCMEC data.
Twitter 2 14
Tagged.com 1 0 NCMEC data also permits analysis of headline

statistics for unique internet protocol (IP)
Pinterest 1 0 addresses used to engage in suspected child
Total 2,645 4,611 exploitation (see Figure 22).
Base: Data provided by AHTCPU.
Figure 22: NCMEC CyberTips concerning suspected child sexual exploitation in Kenya,
number of unique upload IP addresses by year.75
% Change % Change
2017 2018 2019
2017-2019 2018-2019
Kenya Unique Upload IP 8,879 13,027 9,567 8% -27%

Addresses
Total Kenya Reports 12,361 16,108 12,788 3% -21%
Reports per Unique IP Address 1.39 1.24 1.34 -4% 8%
An IP address is assigned to each individual of reports per IP address may be suggestive of a

device on a specific network at a specific time. tendency towards lower volume CSAM offending
Multiple reports per IP address can indicate that within individual online sessions.
suspects (or at least their devices) are engaged in
One foreign law enforcement agency identified
multiple offences of CSAM distribution during the Kenya as a source of commercial forms of live-
same online session, perhaps indicative of a more streaming of child sexual abuse,76 accounting for 2%
deliberate style of offending that is less likely to be of that agency’s reports on this crime type. Another
committed through lack of knowledge. By the same reported sending ten referrals to Kenya regarding
token, Kenya’s consistently low average number OCSEA-related offences in the period 2017-2019.77
75. The same IP address may be counted in more than one year, and a report can contain more than one unique IP address. Technical measures by
ISPs including the dynamic assignment of IP addresses and the sharing of IP version 4 addresses across a large number of devices can also have an
impact on the number of unique IP addresses logged.
76. Also described as ‘live distant child abuse’.
77. INTERPOL requested data and qualitative insights from a number of foreign law enforcement agencies with intelligence on or outreach
activities in the focus countries. In line with intelligence handling protocols and data protection requirements, some of these sources have been
anonymised.
Referrals from foreign law enforcement agencies are Figure 23: CSAM distribution and downloading
most often made when an ongoing investigation is from African Disrupting Harm focus countries,
found to involve an offender or victim in the second observed on peer-to-peer file sharing networks
country, or when a domestic service provider makes by the Child Rescue Coalition.
a report to the national law enforcement authority
that is indicative of OCSEA in the second country. IP Globally Unique
Although the data requirement for this project did Addresses Identifiers (GUIDs)
not include systematic collection of data concerning Ethiopia 7 4
OCSEA referrals from all law enforcement agencies
outside Kenya, it is likely that there have been Kenya 76 24
additional international referrals in the reporting Mozambique 6 10
period, over and above the NCMEC CyberTips
Namibia 94 117
discussed above.
South Africa 2413 842
2.1.3 Evidence of CSAM from other sources
Tanzania 47 5
Hosting: Kenya has not been identified as a hosting
country for images and videos assessed as illegal by Uganda 4 4
INHOPE member hotlines contributing to the ICCAM
Base: Data provided by Child Rescue Coalition for the period of 9th June
platform.78 Moreover, the Internet Watch Foundation 2019 to 8th June 2020.
actioned 0 reports concerning confirmed CSAM
hosting in Kenya in the calendar years 2017, 2018, CSAM distribution on the monitored peer-to-
and 2019. Since data pertaining to the ICCAM project peer networks would appear to be less popular in
is limited to submissions from INHOPE member Kenya than in Southern Africa, but more popular
hotlines, and since the Internet Watch Foundation than in other Eastern African focus countries. In as
operates primarily as the United Kingdom’s CSAM much as data supplied by NCMEC indicates several
hotline, this should not be taken as evidence of an thousand instances of suspected CSAM possession,
absence of CSAM hosting in the country. manufacture and distribution in Kenya in 2017,
2018 and 2019, it would appear that Kenyan CSAM
Distribution on P2P Networks: The Child Rescue offenders may prefer using globally popular US-
Coalition operates the Child Protection System for based platforms to exchange rather than peer-to-
detecting distribution of CSAM on peer-to-peer file peer file-sharing networks.
sharing networks. Data supplied for the time period
9th June 2019 to 8th June 2020 reveals that 76 Web Searches for CSAM: Research was conducted
Kenyan IP addresses were identified as engaged in on Google Trends, with a view to identifying levels
distribution or downloading (see Figure 23). Since the of interest in CSAM in Kenya.79 In the first instance,
system does not monitor all file sharing networks, a sample of 20 terms selected by the INTERPOL
this should not be taken to be representative of the Crimes Against Children team served as keywords
sum total of CSAM offending on such platforms. and phrases for specialist interest in CSAM. Queries
Representation of data for Kenya alongside that for the time period 1 January 2017 to 31 December
for other Disrupting Harm focus countries in Africa 2019 on searches in Kenya returned a result of ’not
allows for comparison. enough data’ for each of these 20 terms.
78. For more information on the ICCAM project, see: INHOPE: What is ICCAM and Why is it Important?.
79. Google Trends (trends.google.com) is a publicly available tool that returns results on the popularity of search terms and strings relative to
others within set parameters. Rather than displaying total search volumes, the tool calculates a score (on a range of 1 to 100) based on a search
term or strings proportion to all searches on all terms/strings. Data points are divided by total searches in the geographical and time parameters
set, to achieve relative popularity. While Google Trends draws on only a sample of Google searches, the dataset is deemed by the company to be
representative given the billions of searches processed per day. For more information on data and scoring, see: FAQ about Google Trends data.
Returns of ‘not enough data’ equate with a 0 relative 2.1.4 Links to travel and tourism
popularity score, indicating a comparatively low level Some data on travelling child sex offenders can also
of interest in that term (as opposed to absolute 0 provide an indication of OCSEA as these offenders
search volume) within the geographical and time often record their sexual abuse or exploitation of
limits set.80 Comparing with global searches for the children for their own use or for further distribution.
same terms and those from other countries in the Online facilitation of CSEA by travelling offenders
same time frame, this suggests that specialist CSAM has also been observed through the use of
search terms may be used less in Kenya than they communications technology to groom or procure
are in some other countries. While it may also be children for offline abuse, or to maintain an online
argued that more sophisticated CSAM searchers are relationship with children whom the offender has
less likely to search on the open web, the relative already abused offline. Among the foreign law
popularity in other countries of some of the terms in enforcement authorities consulted in the context
the Interpol sample would suggest that open web of Disrupting Harm, one agency reported that in
search is still used for CSAM discovery. 2018 they investigated one of their nationals for
Less specialist, more ‘entry level’ searches related to CSEA offences committed in Kenya, while another
CSEA were popular in Kenya in the reporting period, identified the country as a bit of a hotspot for
including English language searches for image and travelling child sex offenders.
video content depicting sexual activity with and In a number of countries, convicted sex offenders
between teenagers, with children, and with babies. are required to notify a central authority of overseas
Related searches for particular formats such as ‘high travel. Analysis of data supplied by one foreign law
definition video’, for material involving children of enforcement agency reveals that 15 notifications
particular ethnicities and for familial abuse appear concerned travel to Kenya between 2015 and 2020,
to indicate that some web searchers in Kenya have representing 0.23% of their total global notifications
specific requirements reflective of a more persistent in that period, and 23.8% of notifications concerning
and active interest in CSAM that has progressed the Disrupting Harm focus countries.81 A second
beyond initial curiosity. foreign law enforcement agency reported that out of
As individuals in Kenya looking for CSAM may 283 notifications of convicted sex offender travel from
search in languages other than English, use of local May 2017 to June 2020, 3% were destined for Kenya.
language and slang search terms present a key In addition, United States Homeland Security
knowledge gap. There is therefore an opportunity Investigations Angel Watch Centre provides referrals
for law enforcement to review OCSEA investigations to officials in destination countries on convicted U.S.
in Kenya, with a view to identifying additional terms child sex offenders who have confirmed scheduled
and search strings used by offenders. The results travel. Those confirmed as not being admitted into
above nevertheless appear to demonstrate that there the destination country are counted as ‘denials’. In
is an appetite for CSAM in Kenya, and the open web the fiscal years 2017 to 2020, the centre made 42
is used for its discovery. referrals concerning travellers to Kenya, representing
29% of the total number of referrals to Disrupting
Harm focus countries in Africa in those years. The
agency received confirmation that nine of these
individuals were denied entry to the country.
80. Ramadanti, D. (2020). Telling stories with Google Trends using Pytrends in Python.
81. INTERPOL requested data and qualitative insights from a number of foreign law enforcement agencies with intelligence on or outreach
activities in the focus countries. In line with intelligence handling protocols and data protection requirements, some of these sources have been
anonymised.
2.2 CHILDREN'S EXPERIENCES OF CHILD SEXUAL
EXPLOITATION AND ABUSE IN KENYA
Under the Disrupting Harm project, OCSEA was defined specifically to include
CSAM, live-streaming of child sexual abuse and online grooming of children for
sexual purposes. These concepts are used here to organise and present the results
of our research. At the same time, we recognise that the ways in which children are
subjected to OCSEA are far more complex and nuanced. The experiences or offences
in question often occur in combination or in sequence. Moreover, as explored in the
box The Continuum of Online and Offline Child Sexual Exploitation and Abuse on
page 63 OCSEA does not only occur in the digital environment; digital technology can
also be used as a tool to facilitate or record in-person sexual exploitation and abuse.
2.2.1 Online grooming a child to sexualised content and the Sexual

Disrupting Harm defines online grooming as Offences Act where we use the child pornography
engaging a child via technology with the intent of section. But we cannot directly charge grooming
sexually abusing or exploiting the child. This may as an offence.” (RA4-KY-08-A-justice) However, an
happen either completely online or a combination of interview with a representative from the Kenya Law
online and in person. Reform Commission confirmed that Section 20 (3)
of the upcoming Children Bill 2021 will expressly
Online grooming is a complex process which is criminalise proposing to meet a child for sexual
often fluid and difficult to detect, especially where it purposes through electronic systems, networks or
involves a slow build of trust between the offender communication technologies. Convicted offenders
and the child over an extended period of time. The would be liable to imprisonment not exceeding ten
child is often ‘prepared’ for sexual abuse and made years or a fine not exceeding two million shillings
to engage in sexual acts online or in person by (approximately US$18,000) or both. (RA1-KY-09-B)
means of deceit, coercion or threats. However, online
grooming can also be abrupt, with an offender The definition contained in the upcoming Children
suddenly requesting or pressuring a child to share Bill 2021 will be in line with the international
sexual content of themselves or to engage in sexual standard established by the Council of Europe’s
acts, including via extortion. Of the nine Kenyan girls Convention on the Protection of Children Against
which Disrupting Harm experts spoke to as part of Sexual Exploitation and Sexual Abuse (Lanzarote
the survivor conversations research activity, all had Convention)82 and the EU Directive 2011/93,83 the
experienced online grooming as a part of the sexual only two legally binding international instruments
exploitation and abuse they were subjected to. containing an obligation to criminalise the grooming
of children for sexual purposes. However, it has been
Legislation on grooming noted that these definitions themselves warrant
At the time of writing, Kenyan law does not updates as they require an intention to meet the
specifically criminalise the grooming of children child in person. In 2015 the Lanzarote Committee
for sexual purposes. An investigator from the issued an opinion recommending that states should
AHTCPU highlighted this in one of our access to extend the crime of grooming for sexual purposes to
justice interviews, saying that they “rely on the include “cases when the sexual abuse is not the result
Computer Misuse Act which penalises exposing of a meeting in person, but is committed online.” 84
82. Council of Europe. (2007). Convention on the Protection of Children against Sexual Exploitation and Sexual Abuse. Council of Europe Treaty
Series – No. 201. Article 23.
83. European Parliament and of the Council. (2011). Directive 2011/92/EU on combating the sexual abuse and sexual exploitation of children and
child pornography, and replacing Council Framework Decision 2004/68/JHA. Article 6.
84. Council of Europe’s Lanzarote Committee. (2015). Opinion on Article 23 of the Lanzarote Convention and its explanatory note. Para 20.
In Kenya, the provision of the upcoming Children year. There were no notable differences by gender
Bill 2021 addressing online grooming will only cover whereas 16-17 year olds were somewhat more likely
online grooming with the intent of meeting the child to receive this requests compared to 12-13 year
and may not therefore apply to situations where, for olds (15% and 10%, respectively). Depending on
example, a child is asked to send sexual content to an the context, these experiences could mean varying
offender via online platforms. The bill received cabinet levels of harm for a child. For example, a child being
approval on 25 February 2021 but as of early June 2021 asked to talk about sex by a boyfriend or girlfriend
it still required three readings in Parliament before but not wanting to engage at that moment might
going to the President for promulgation. It is hoped not face serious harm from this interaction. On the
that a provision outlawing grooming where the sexual other hand, these experiences could also indicate
abuse occurs online could still be included. malicious instances of attempted grooming;
therefore, we report on it here and describe the
Patterns of victimisation figures above as instances of potential (versus actual)
Although Kenya lacks law enforcement data on the grooming.
modus operandi of grooming, the conversations with
survivors of OCSEA provided insights into the tactics Online or offline? One third of the children who
used by offenders to commit offences and ensure were asked to talk about sex when they did not want
victim compliance is achieved. In our conversations to, received these requests in person, 37% on social
with nine survivors of OCSEA from Kenya, a pattern media and 10% in an online game. The 47 children
of flattery, offers of money and other goods emerged. who were asked to talk about sex via social media
For example: were most likely to say it happened on Facebook or
Facebook Messenger, WhatsApp and YouTube.
• “…the kind of things every young person wants to
hear, like ‘You are very beautiful’.” (RA5-KY-02-A) Because asking a child to talk about sex can happen
• “Yes, just those flattering words that men use, like without the involvement of technology, only children
‘You are beautiful’ and ‘I’ll buy you a present’. He who most recently received these requests on social
once sent me money through my aunt’s number media or in an online game (58 of the 127 children)
and she questioned me but I just told her to give were included in the subsequent analysis, as they
me the money. It was after sending me the money represent potential OCSEA cases.
that we met the week later.” (RA5-KY-09-A)
How children felt: Of the 58 children who were
• “After a while he started making promises like he’ll asked to talk about sex when they did not want to
pay my fees [school fees] and even give me pocket via an online channel (i.e., social media or an online
money any time I needed. So I was lured with game), 75% reported negative feelings about the
money and decided to finally meet him.” (RA5- experience, while one in four said it did not affect
KY-07) them. The most common negative feelings cited
by these children were feelings of embarrassment
The men these children met were all older than
or annoyance. Other children said they felt angry,
them:
betrayed, guilty, distressed and scared.
• “He didn’t live round here and he was way older
How children respond: In our subsample of 58
than me.” (RA5-KY-01-A)
children who were asked online to talk about sex
• “At first I thought we were age mates considering when they did not want to, 36% refused to do so.
he did not post his pictures but just quotes. I was About a quarter blocked the offender, while 17% of
even shocked when we met...” (RA5-KY-07-A) children ignored the problem and hoped it would go
Potential grooming – children asked to talk away on its own. Others asked the offender to leave
about sex them alone, stopped using the internet for a while,
Of the participants in our household survey of 1,014 changed privacy settings, and deleted messages
internet-using 12-17-year-olds in Kenya, 13% (127 from the sender. Nevertheless, a small proportion
children) had been asked to talk about sex or sexual (4 children) did as they were asked.
acts when they did not want to within the past
IN THE PAST YEAR
YES 13%
I HAVE BEEN ASKED TO TALK ABOUT
n= 1,014 children
SEX WHEN I DID NOT WANT TO

THE LAST TIME THIS HAPPENED ONLINE…
What did you do?*† How did you feel?* Who did it?*†
Annoyed A friend/acquaintance (18+)

36% 28%
A romantic partner (or ex-)
14% 26%
24% A friend/acquaintance (under 18)

23% 25% 16%
17% A family member

9%
Blocked the
Prefer not to say

Embarrassed It didn't
Ignored it
9%
Said no
affect me
person
Someone unknown to the child

31%
n= 58 internet-using children aged 12-17 who received unwanted requests online to talk about sex in the past year.
Where did it happen?*† Whom did you tell?**†

No one
Friend
38%
37% 33%
10% 14%
37%
In an online game
Some other way
TOP 3
Social media
12%
In person
Sibling
BOTTOM 3
n= 127 internet-using children aged 12-17

who received unwanted requests to talk 0% 0% 0%
about sex in the past year.
Police Helpline Social worker
n= 58 internet-using children aged 12-17 who received

On which platform did this happen?*† unwanted requests online to talk about sex in the past year.
Why did you not tell anyone?*†

Facebook or
Facebook Messenger WhatsApp YouTube
I did not think
I did not know anyone would I felt that I did
whom to tell believe me something wrong
49% 48% 21%
27% 23% 22%

n= 47 internet-using children aged 12-17 who most recently
received unwanted requests via social media to talk about sex.
n= 22 internet-using children aged 12-17 who did not tell
anyone the last time they received unwanted requests online
to talk about sex.
*These figures represent the most common responses selected by children.
**These figures represent the most and least common responses selected by children.
†
Multiple choice question Source: Disrupting Harm data
2.2 CHILDREN'S EXPERIENCES OF CHILD SEXUAL EXPLOITATION AND ABUSE IN KENYA
Who makes the requests: As illustrated on

page 46, the people who most commonly asked
children to talk about sex online were adult friends Most children who received
or acquaintances, followed by current or former unwanted requests to talk
romantic partners, friends or acquaintances aged
under 18, and family members. Close to one third of about sex online did not
children said the offender was someone unknown to
confide in anyone all. Some
them. Although this is not a small proportion of cases,
when taken together, those who are already part of of them told a friend, but few
the child’s life are more likely to send them these
kinds of requests than individuals unknown to the
spoke to an adult about it.
child.
Whom children tell about it – if anyone: Twenty-two

children who received unwanted requests to talk about
How children respond: Of the 101 children who
sex online did not confide in anyone at all. Children
received unwanted requests to share images of their
who did disclose what happened were most likely to
private parts, 44% refused. Other common responses
tell a friend, but few spoke to an adult about it.
included blocking the other person, ignoring the
None of the 58 children reported what had problem and hoping it would go away by itself. Eleven
happened to them through an online reporting percent of children changed their privacy settings.
function. This could be because children do not
Nine percent of children who were asked to share
know where to go or whom to tell about these
sexual images or videos of themselves complied.
experiences. This was the most common reason
The youngest children (aged 12-13) were most likely
provided by children who did not talk to anyone
to comply. Boys were more likely than girls to agree
about what happened. In fact, only 24% of the
to send images or videos of their private parts even
children in our full sample of 1,014 were confident
though they did not want to.
that they knew how to report harmful content on
social media, while 61% said they did not know Who makes the requests: Of the children who had
where to get help if they or a friend were subjected received unwanted requests to share a sexual image
to sexual harassment or abuse. or video of themselves, 32% named a romantic
partner as the source of the request. Adult friends or
Potential grooming – children asked to share
acquaintances accounted for 23% of the cases, a family
sexual images or videos
member for 15% and a friend younger than 18 for
Some offenders have the intention of manipulating
12%. A quarter of the children said the offender was a
children into self-generating and sharing sexual images
someone unknown to them.
or videos through digital technologies, whether or not
they also intend to meet the child in person. Overall, children are more likely to receive unwanted
requests to talk about sex or share sexual content by
A behaviour that could be an indication of grooming
people they already know, rather than by individuals
is sending children unwanted requests to share
unknown to them. The fact that at least one request
sexual content of themselves. Within the past year,
in five came from an adult makes it likely that some of
10% of the internet-using children we surveyed
these experiences constituted grooming.
(101 children) had received unwanted requests
for a photo or video showing their private parts. Online or offline? More than half (54%) of the 101
There were only minor differences by gender or children who were subjected to unwanted requests
age group. While 20% of these children said they for sexual content said the requests were made via
were not affected the last time they received such social media. Seven percent of children said the
a request, 76% felt negatively about it. Feelings of requests came through online games. About one
embarrassment and anger were the most common, in five said they were asked in person. Once again,
followed by being annoyed, scared or distressed.
IN THE PAST YEAR
YES 10%
I WAS ASKED FOR A PHOTO OR VIDEO
n= 1,014 children
SHOWING MY PRIVATE PARTS WHEN

I DID NOT WANT TO
THE LAST TIME THIS HAPPENED…
What did you do?*† How did you feel?* Who did it?*†
44%
Angry A romantic partner (or ex-)
32%
A friend/acquaintance (18+)
14% 23%
23% 20% A family member
15%
A friend/acquaintance (under 18)
15% 13% Embarrassed 12%
It didn't
Prefer not to say
the person
affect me
6%
Blocked
Said no
Ignored

it
25%
n = 101 internet-using children aged 12-17 who received unwanted requests for sexual images in the past year.

No one
31%
54% 21%
7% 13% Friend
40%
In an online game
Some other way
TOP 3
Social media
7%
In person
Male caregiver
BOTTOM 3
n= 101 internet-using children aged

12-17 who received unwanted requests 2% 2% 1%
for sexual images in the past year.
On which platform did this happen?*† n= 101 internet-using children aged 12-17 who received
unwanted requests for sexual images in the past year.

Facebook or
WhatsApp Facebook Messenger Instagram
I worried I
would get I felt I did not know
embarrassed whom to tell
58% 53%
in trouble
17%
32% 26% 23%

received unwanted requests for sexual images via social media.
anyone the last time they received unwanted requests for
**These figures represent the most and least common responses selected by children. sexual images.
†
Multiple choice question Source: Disrupting Harm data
WhatsApp and Facebook or Facebook Messenger Over a third of the children offered money or gifts in
were the social media and instant messaging apps return for sexual images told a friend the last time this
via which children were most commonly targeted. happened. Caregivers were the next most common
This is probably because Facebook and WhatsApp confidants – 16% told a male caregiver and 10% told a
– the two most popular social media platforms in female caregiver. Only one of the 67 children spoke to a
Kenya85 – are where children spend much of their helpline and no one reported to the police or spoke to a
time online. social worker Twenty-eight percent did not tell anyone at
all. The 19 children who did not disclose or report what
Whom children tell about it – if anyone: Among happened said that the main barrier to reporting was
children who received unwanted requests to send not knowing where to go or whom to tell. Other children
images or videos showing their private parts, 40% told said that they would feel embarrassed or ashamed or
a friend, but almost one in three did not share their that it would be emotionally difficult for them to share
experience with anyone. Very few survey respondents their experiences.
formally reported what happened to them through an
online reporting system. For the 31 children who did Offering children money or gifts for sexual acts in
not disclose the unwanted requests for sexual content person
to anyone, the main barriers were being worried It is clear from the conversations with survivors of
about getting into trouble, feeling embarrassed, or not OCSEA conducted as part of the research for
knowing where to go or whom to tell. Disrupting Harm that the grooming of children online
for the purpose of meeting in person to engage in
Offering children money or gifts for sexual images sexual activities can be a real threat.
or videos
The offer of money or gifts to a child in return for All nine of the Kenyan girls we spoke to in our OCSEA
sexual images or videos constitutes evidence of survivor conversations had met an offender in person
grooming with the aim of obtaining CSAM. Seven after connecting online and then been subjected to
percent of children in the household survey (67 sexual abuse and exploitation. Of note is that none of
children) said they had been offered money or gifts these children spoke of sexual images or videos being
in return for sexual images or videos in the past year. made or exchanged online. While this sample is not
There were no clear differences by age group or representative of all survivors, it provides some insight
gender. into the interactions between children and offenders.
Asked about the last time they were offered money • “Yes. When I told him I was going home since it was
or gifts in exchange for sexual content, most of late, he said he has to have sex with me. So when I
the 67 children said they received the offer from wanted to escape he forced me.” (RA5-KY-05-A)
someone they already knew. Close to a quarter were
offered money or gifts by romantic partners. These To return to our household survey results with 1,014
were followed by friends or acquaintances younger children, 7% of the internet-using children surveyed
than 18, and adult friends or acquaintances. Among (67 children) said they had been offered money or
persons known to the child, family members were gifts to meet someone in person to do something
the least likely to make offers of this kind (12%) (see sexual within the past year. Like other findings, these
page 50 below). Individuals unknown to the child numbers may be under-reported as children may
accounted for around one third of all cases. not feel comfortable or safe enough to disclose their
experiences of abuse and exploitation.
While 25% of the children said that the offer of
money or gifts was made in person, most offers were Online or offline? Of the 67 children who said they
made online – 49% on social media and 18% via an had been offered money or gifts to meet in person
online game. Among the 33 children who received for sexual activities in the past year, almost half said
such offers via social media, the most common that this unwanted request came through social
platforms cited were Facebook or Facebook media and 9% through an online game (see page 51).
Messenger, WhatsApp and YouTube. Four children Nineteen percent of children received the offer
cited Instagram and two TikTok. in person.
85. United States International University – Africa. (2019). Social Media Consumption in Kenya: Trends and Practices.
OCSEA
YES 7%
IN THE PAST YEAR
I WAS OFFERED MONEY OR GIFTS IN

n= 1,014 children
RETURN FOR SEXUAL IMAGES OR VIDEOS

Who did it?*†
31%
24%
21% 21%
12%
8%
A romantic A friend/ A friend/ A family Prefer not Someone

partner (or ex-) acquaintance acquaintance member to say unknown to
(under 18) (18+) the child
n= 67 internet-using children aged 12-17 who were offered money or gifts for sexual images or videos.

No one
28%
49% 25%
Friend
18% 9%
34%
Some other way
TOP 3
Social media
In an online
16%
In person
game
Male caregiver
BOTTOM 3
12-17 who were offered money or gifts

for sexual images or videos. 0% 2% 0%
n= 67 internet-using children aged 12-17 who were

On which platform did this happen?*† offered money or gifts for sexual images or videos.

Facebook or
Facebook Messenger WhatsApp YouTube
I did not think
I did not know anyone would
whom to tell I felt embarrassed believe me
58% 47% 24%
47% 37% 26%

were offered money or gifts via social media in exchange for
sexual images or videos.
anyone the last time they were offered money or gifts for sexual
images or videos.
50 †
Multiple choice question Disrupting Harm in Kenya – Evidence on online child sexual exploitation andHarm
Source: Disrupting abuse
data
OCSEA
YES 7%
IN THE PAST YEAR
I WAS OFFERED MONEY OR GIFTS TO MEET

n= 1,014 children
IN PERSON TO DO SOMETHING SEXUAL

Who did it?*†
32% 31%
21%
18%
16%
5%
A friend/ A friend/ A romantic A family Prefer not Someone

acquaintance acquaintance partner (or ex-) member to say unknown to
n= 38 internet-using children aged 12-17 who were offered money or gifts online for in-person sexual acts in the past year.

No one
32%
49% 19% 18%
9% Friend
40%
In an online game
Some other way
TOP 3
Social media
15%
In person
Sibling
BOTTOM 3
who were offered money or gifts for 5%

in-person sexual acts in the past year. 0% 0%
Helpline Police Social worker

On which platform did this happen?*† offered money or gifts online for in-person sexual acts in
the past year.

Facebook or
Facebook Messenger WhatsApp Instagram
I did not know I felt I worried I would

whom to tell embarrassed get in trouble
58% 56% 19%
42% 42% 25%

received offers of money or gifts for in-person sexual acts via
social media.
anyone the last time they were offered money or gifts online
*These figures represent the most common responses selected by children. for in-person sexual acts.
Disrupting
† Harm
Multiple choice in Kenya – Evidence on online child sexual exploitation and abuse
question Source: Disrupting Harm data 51
Among the 33 children who received offers of money

or gifts to engage in sexual acts in person via social
media, the most common platforms cited were Seven percent of the
Facebook or Facebook Messenger and WhatsApp, internet-using children in the
followed by Instagram.
household survey said that
Among the 38 children in the survey who had been
offered money or gifts online – i.e., via social media they had been threatened
or an online game – to meet in person for sexual or blackmailed to engage in
acts, the offers came from a range of sources. These
included offers from someone unknown to the child, sexual activities within the
from a peer younger than 18 and from an adult friend past year.
or acquaintance. Current or former romantic partners
and family members were less likely to make offers of
this kind.
Once again, children were very unlikely to report these last time this happened they had been threatened
incidents through formal channels and instead tend to or blackmailed via social media and 9% through an
confide in the people close to them, most commonly a online game. However, it also happened in person to
friend or sibling. Almost a third of the children who had a considerable extent. Among social media channels,
been offered money or gifts in return for sexual acts the most common platforms where children
via online channels (12 children) did not tell anyone. experienced this were Facebook or Facebook
The most common reasons which these children gave Messenger and WhatsApp. Almost one third of these
for not disclosing their experiences were feelings of children cited Instagram. Twenty-eight percent of
embarrassment and shame (explored in more detail the children said that they had been threatened or
in the box ‘Social and cultural barriers to disclosing blackmailed in person.
OCSEA in Kenya’ on page 67) and not knowing where
to go or whom to tell. For the 38 children who were threatened or
blackmailed online – i.e., via social media or an online
Sexual extortion game – the most common offender was an adult
Sexual extortion is sometimes used in the grooming friend or other acquaintance, followed by someone
process. Often the offenders have already obtained unknown to the child, and a friend or acquaintance
sexual images of the children and threaten to younger than 18 years. Seven children reported that
publicly publish or share these with their friends or current or former romantic partners did it, and three
family members as a way of coercing children into said that it was done by a family member. Overall,
sharing more images or engaging in other kinds of as with the other forms of OCSEA explored in this
sexual activities. Such threats can also be used to chapter, sexual extortion was more commonly
extort money. In Kenya, sexual extortion committed committed by individuals known to the child than by
online is not specifically criminalised by law. people they do not know.
Seven percent of the internet-using children in the Of the 38 children, 18 told a friend about the incident,
household survey (71 children) said that they had whereas nine did not tell anyone. Low disclosure
been threatened or blackmailed to engage in sexual of sexual extortion is perhaps to be expected as
activities within the past year. It is unclear what kind it is based on threatening to disclose images and
of threats were used. No question was asked about cause embarrassment, making it even harder to
the use of sexual images to extort money. seek help. Only one child reported to the police, one
child spoke to a social worker and one child called a
The use of online channels was common for this helpline.
kind of abuse. Of the 71 children, 45% said that the
OCSEA
YES 7%
IN THE PAST YEAR
SOMEONE THREATENED OR BLACKMAILED

n= 1,014 children
ME TO ENGAGE IN SEXUAL ACTIVITIES

Who did it?*†
34%
26%
23%
18%
8%
A friend/ A friend/ A romantic A family Someone

acquaintance acquaintance partner (or ex-) member unknown to
(18+) (under 18) the child
n= 38 internet-using children aged 12-17 who were threatened or blackmailed online to engage in sexual acts in the past year.

No one
Friend
24%
45% 28% 9% 17%
47%
In an online game
Some other way
10%
TOP 3
Social media
In person
Male caregiver

BOTTOM 3
3% 3% 3%
who were threatened or blackmailed to
engage in sexual acts in the past year.
Helpline Police Social worker
On which platform did this happen?*† n= 38 Internet-using children aged 12-17 who were
threatened or blackmailed online to engage in sexual
acts in the past year.
Facebook or Why did you not tell anyone?*†

Facebook Messenger WhatsApp Instagram
I did not
I worried I think anyone
I did not know I felt would get would
66%
whom to tell embarrassed in trouble believe me
47% 30%
56% 33% 22% 22%

received threats or were blackmailed via social media.
n= 9 internet-using children aged 12-17 who did not tell anyone the
last time they were threatened or blackmailed online to engage in
*These figures represent the most common responses selected by children. sexual activities.
Disrupting
†
Harm
The most common reasons given by the children streaming is not defined in our law; so prosecuting
who did not disclose the incident were not knowing such a case is difficult because it’s not anchored and is
where to go or whom to tell, feeling embarrassed, not defined in any law.” (RA1-KY-08-A)
being worried about getting in trouble and not
thinking anyone would believe them. The upcoming Children Bill 2021 will not expressly
define this crime either. However, a senior researcher
2.2.2 CSAM and live-streaming of child from the Kenya Law Reform Commission argued that
sexual abuse articles 20(3)(b) and (c) of the Act would implicitly
Kenyan legislation explicitly defines CSAM and cover live-streaming of child sexual abuse as they
criminalises acts associated with it.86,87 The legal refer to “transmission of obscene material” and
definition covers visual and audio material as “online abuse and exploitation.” (RA1-KY-09-B) While
well as digitally-generated CSAM.88 However, CSAM and live-streaming of child sexual abuse are
knowingly obtaining access to CSAM is not currently considered separate concepts in law, the
explicitly criminalised – a major loophole. Similarly, distinction is artificial because live-streaming of
live-streaming of child sexual abuse is not an child sexual abuse is one way in which CSAM can be
explicit offence in current legislation. This can pose produced, disseminated and consumed. Legislators
challenges for prosecution. A respondent in one of our should be aware of the overlaps between these
government duty-bearer interviews stated that “Live- concepts.
How technological development has influenced Concerningly for law enforcement authorities,
OCSEA many streaming platforms do not create any
The wide availability of faster and cheaper internet records, because video is not downloaded or
access has led to the increasing use of video tools retained by default, although metadata is. This
in communications. Video chat and live-streaming means that when the streaming stops the CSAM
tools have rapidly gained popularity and are vanishes, unless the offender deliberately records
changing the ways we engage with each other, it. This increases the chances of impunity for
particularly for young people (34% of 12-17 internet offenders, and creates specific challenges for
users in Kenya watch live-streams weekly). Live- investigators, prosecutors and courts, especially
streaming is increasingly used both amongst small as the existing legal definitions of CSAM and
private groups and for ‘broadcasts’ to large, public, methods of investigation and prosecution are not
unknown audiences. While this is often harmless always up to date.
and has many benefits, the misuse of such tools is
creating new ways of perpetrating OCSEA. Self-generated sexual content involving children:
As noted in chapter 1.3.3, the rise in self-generated
Offenders broadcasting child sexual abuse: sexual content, both coerced and non-coerced, also
Live-streaming tools can be used to transmit includes live-streaming. This content poses complex
sexual abuse of children instantaneously to one challenges. Even if its production is non-coerced,
or more viewers, so that they can watch it while this content may still make its way into circulation
it is taking place. Remote viewers may even be through non-consensual on-sharing or nefarious
able to request and direct the abuse, and financial means, such as hacking. Governments and support
transactions can occur alongside it or even within services everywhere are grappling with how to
the same platforms. address these issues.
86. Republic of Kenya. (2018). The Computer Misuse and Cybercrimes Act No. 5 of 2018. Section 24.
87. Republic of Kenya. (2006). The Sexual Offences Act No. 3 of 2006. Section 16 (3)(b). (Last revised in 2019).
88. Republic of Kenya. (2018). The Computer Misuse and Cybercrimes Act No. 5 of 2018. Section 24 (3).
Children’s experiences of non-consensual
sharing of sexual images
Data from the NCMEC’s CyberTipline presented in “In one of the cases we were
chapter 2.1 show that the possession, manufacture dealing with, the caregivers were
and distribution of CSAM accounted for almost all of
Kenya’s NCMEC CyberTips in 2017-2019.
feeling like a case involving CSAM
was not serious as it only involved
Moreover, 7% of the internet-using children aged
12-17 in Kenya (72 children) who took part in the a picture so their attitude was
Disrupting Harm household survey stated that ‘This child is only on the picture,
someone had shared sexual images of them
without their permission in the past year, with no no one has done anything to her,
notable variations by gender or age group.89 This is so why do you want us to pursue
an alarming number considering the severity of this
crime. These images, and particularly those shared a court case?’”
online, can be circulated widely and repeatedly
viewed all over the world, resulting in a continuous
sense of fear of being recognised for the victims.
When these images or videos are recordings of According to the 72 children in the household survey
severe sexual abuse, the trauma associated with whose sexual images had been shared without their
those in-person experiences can also be repeatedly permission the last time this happened to them, the
reactivated as the content is shared further. persons most commonly responsible were individuals
unknown to the child, friends or acquaintances
younger than 18, adult friends or acquaintances,
Case Study – High school students’ and (current or former) romantic partners. Family
experiences of non-consensual sharing members were least likely to share sexual images
of sexual images without permission.
In April 2020, Kenya’s Directorate of Criminal
Fifty-five percent of the children whose images had
Investigations received a call from one of
been shared without their permission said that
the banks in the country about a suspected
the they were shared via social media – particularly
case of non-consensual sharing of sexual
WhatsApp, followed by Facebook/Facebook Messenger
images. The case concerned a student with a
and YouTube. Boys were more likely to have their
scholarship from the bank, an 18-year-old who
images shared on social media than girls. Only 4%
had reported to the police that his schoolmate,
said the images were shared through an online game.
a 17-year-old boy, had circulated the victim’s
Fourteen percent said they were shared in person.
nude photos via social media (WhatsApp,
Instagram and Twitter) under fake accounts The children abused in this way were most likely
purported to belong to the victim. According to confide in a friend or not to tell anyone at all.
to the victim, the suspect had posed as a girl As shown on page 56, a few children confided in
on social media and tricked him into sharing a sibling or caregiver. Almost no one turned to a
the images. The accounts showing his images helpline, and no children reported the incident to
also portrayed him as gay (which is illegal in the police or a social worker. Among the 22 children
Kenya). The Directorate requested Instagram, who did not tell anyone, the most common reasons
Twitter and Facebook to preserve the evidence for not disclosing were not knowing where to go
and pull down the fake accounts. The case is or whom to tell, worries over getting into trouble,
Pending Under Investigation. embarrassment and not thinking anyone would
believe them.
89. Note that all sexual images of a child are in fact defined as CSAM under Kenyan law, regardless of whether the child shares them voluntarily or
they are shared without permission.
OCSEA
YES 7%
IN THE PAST YEAR
someone shared sexual images

n= 1,014 children
of me without my permission
39%
Who did it?*†
27%
21%
18%
8%
3%
A friend/ A friend/ A romantic A family Prefer not Someone

acquaintance acquaintance partner (or ex-) member to say unknown to
n= 72 internet-using children aged 12-17 whose sexual images were shared non-consensually in the past year.

No one
4%
31%
55%
14% 17%
Friend
38%
In an online game
Some other way
TOP 3
Social media
9%
In person
Sibling
BOTTOM 3
n= 72 internet-using children aged 12-17 4%

whose sexual images were shared 0% 0%
non-consensually in the past year.
n= 72 internet-using children aged 12-17 whose sexual

images were shared non-consensually in the past year.
On which platform did this happen?*†

Facebook or
WhatsApp Facebook Messenger YouTube
I did not know I was worried I felt I did not think

whom to tell I would get in embarrassed anyone would
56% 41%
trouble believe me
13%
36% 36% 18% 18%

n= 39 internet-using children aged 12-17 whose sexual images were
most recently shared via social media.
n= 22 Internet-using children aged 12-17 who did not tell anyone the
last time their sexual images were shared non-consensually.

56 †
Multiple choice question Disrupting Harm in Kenya – Evidence on online child sexual exploitation and abuse
Attitudes to non-consensual sharing

of sexual images
Not everybody is aware of the gravity of sharing This suggests that one out of
sexual images of others without their permission. every twenty internet users in
According to one justice professional interviewed for
this report: “In one of the cases we were dealing with,
Kenya in this age group may
the caregivers were feeling like a case involving CSAM receive money for sexual images
was not serious as it only involved a picture so their
attitude was ‘This child is only on the picture, no one
or videos at least once a year.
has done anything to her, so why do you want us to
pursue a court case?’” (RA4-KY-05-A-justice)
The findings of our household survey of 1,014

internet-using 12-17-year-olds and their caregivers were ‘harvested’ from the original online location and
show a degree of awareness. Sixty-seven percent shared on third party websites.93
of the children and 83% of the caregivers surveyed
Given the sensitivity of this topic, only the 15–17-year-
agreed that if a person has naked images or videos
old respondents in the household survey were
of someone else, it should be illegal to share them
asked whether they had accepted money or gifts in
with other people. Among the same children and
exchange for sexual images or videos of themselves.
caregivers, however, 63% and 74% respectively
Among the 563 respondents who were asked, 5% said
were of the opinion that “if someone takes naked
they had done this in the past year. This suggests that
images or videos of themselves, it is their fault if
one out of every twenty internet users in Kenya in this
they are shared with other people”. This kind of
age group may receive money for sexual images or
victim-blaming may partly explain the low levels of
videos at least once a year. Some children may have
reporting by children subjected to various forms of
been hesitant to reveal their involvement in such
OCSEA in the past year.
activities – even in an anonymised survey – so the true
Accepting money or gifts in exchange for sexual figure could be even higher.
images or videos
By making financial micro-transactions easy and
As we explored in the context of grooming, children
instant, the growing use of digital and mobile
are sometimes offered money or gifts in return for
payments may facilitate this form of OCSEA. Kenya is
sexual content. Here we consider the acceptance
known for its widespread use of digital payments.94,95
of money or gifts by children in return for sexual
As of March 2019, 223,084 mobile money agents and
content, regardless of how the process was initiated.
32 million subscriptions to mobile money transfers
While the practice of accepting money or gifts in were recorded in the country. More than 80% of
exchange for sexual activities is not new,90,91,92 the use of the mobile money transactions were referred to
digital technologies – including by children and young M-Pesa – the first mobile money service launched in
people – to self-produce and send images or videos of 2007, which offers retail financial services via mobile
oneself in return for money or other material incentives phones to Kenyans, especially those in under-served
is an emerging trend. This practice could increase the rural areas.96 Globally, there has long been concern,
risk of non-consensual sharing: 90% of the ‘youth- particularly within the law enforcement community,
generated’ sexual images and videos assessed in a of the risk of ‘borderless’ cryptocurrencies being
study by the Internet Watch Foundation and Microsoft misused to facilitate child abuse.97
90. Zulu, E.M. , F.N. Dodoo and A.C. Ezeh (2002). “Sexual Risk-Taking in the Slums of Nairobi, Kenya, 1993-98.” Population Studies 56(3):311-323.
91. Kabiru, C. W., Beguy, D., Undie, C.-C., Zulu, E. M., & Ezeh, A. C. (2010). Transition into first sex among adolescents in slum and non-slum
communities in Nairobi, Kenya. Journal of Youth Studies, 13(4), 453–471.
92. Stoebenau, K., Heise, L.,Wamoyi, J., & Bobrova, N. (2016). Revisiting the understanding of “transactional sex” in sub-Saharan Africa: A review and
synthesis of the literature. Social Science & Medicine, vol. 168, 186-197.
93. Internet Watch Foundation & Microsoft. (2015). Emerging Patterns and Trends Report #1 Online-Produced Sexual Content.
94. The Kenyan Wall Street. (2019). Kenya Tops in Mobile Money Penetration Globally.
95. World Bank. (2018). The Little Book on Financial Inclusion.
96. World Bank. (2018). What Kenya’s mobile money success could mean for the Arab world.
97. Internet Watch Foundation. (2014). Briefing Paper – Preliminary Analysis of New Commercial CSAM Website Accepting Payment by Bitcoin.
Caregivers’ knowledge about OCSEA
Six of the ten caregivers of young people who had • “It is where children are abused online.”
accessed the justice system who were interviewed (RA4-KY-06B-caregiver and RA4-KY-09B-caregiver)
for Disrupting Harm said that they had previous • “…things that happen to children when they are
knowledge of OCSEA. The other four only became exposed to things online and when there’s no
aware of the phenomenon after children in their limitations or supervision by their caregivers”
care were abused. Although they said that they (RA4-KY-05B-caregiver)
now know what OCSEA is, when asked to describe
it, most of the caregivers gave limited explanations: According to our household survey of internet-using
children and their caregivers, caregivers in Kenya are
• “It is an exploitation that happens online for most likely to obtain information from family or friends,
children on the internet they are exposed to for religious leaders, television, and schools on how to keep
the purpose of exploitation or abuse.” (RA4-KY- their children safe online (see Figure 24).
10B-caregiver)
Figure 24: Caregivers’ sources of information on how to support their children’s internet use
and keep them safe online
Family or friends
32%
Religious leaders
31%
Television
28%
Child's school
26%
Radio
21%
Social media
21%
Don't get any information about this

20%
Online safety course

8%
Newspapers or brochures
4%
Other sources
2%
0% 10% 20% 30% 40% 50%
Base: Caregivers of internet-using children aged 12-17 in Kenya (For ‘online safety course’ and ‘social media’: Caregivers who use the internet).
n = 1,014 (For ‘online safety course’ and ‘social media’, n = 522).
Figure 25 shows the channels through which broadly coincide with their existing sources of
caregivers would ideally like to obtain information information, religious leaders and the child’s
on the same subject. While these channels school are the two most popular answers.
Figure 25: Caregivers’ preferred sources of information on how to support their children’s
internet use and keep them safe online
Religious leaders
41%
Child's school
38%
Television
35%
Family or friends
31%
Radio
26%
Social media
15%
Online safety course

10%
Other sources
5%
Newspapers or brochure
3%
0% 10% 20% 30% 40% 50%
2.3 OTHER EXPERIENCES OF CHILDREN THAT MAY BE
LINKED TO OCSEA
Additional to the examples of OCSEA already presented, children may be subject
to other experiences online which can be harmful, such as sexual harassment or
unwanted exposure to sexualised content. Moreover, these experiences could,
in some instances, contribute to the desensitisation of children so that they
become more likely to engage in sexual talk or sexual acts – for example, during a
grooming process.
2.3.1 Sexual harassment than 18 and an adult friend or acquaintance. Around one
Kenyan legislation does not explicitly criminalise online in four children said the offender was someone they did
sexual harassment of children. However, section 27 of not know. The youngest respondents were most likely to
the Computer Misuse and Cybercrimes Act makes receive these comments from a family member (24%)
‘cyber harassment’ an offence. In view of the broad compared to 5% of 14-15-year-olds and 8% of 16-17-year-
wording used, this provision could be invoked for sexual olds. Older children were more likely to be targeted by
harassment of children online. Duty-bearer interviews strangers (16-17: 31% vs. 12-13: 20%).
with representatives from the Kenya Law Reform As with other forms of sexual violence, most children
Commission also confirmed that the concept of either told a friend or did not tell anyone at all the last
online abuse referred to in subsection 20(3)(c) of time they were subjected to this kind of harassment.
the upcoming Children Bill 2021 will encompass Among the 74 children who did not tell anyone the
cyber harassment and cyber bullying, among last time this happened to them, the most common
other phenomena. barriers were not knowing where to go or whom to tell,
Our household survey of 12-17-year-olds shows that in and feeling embarrassed or ashamed or that it would be
the past year, 21% internet-using children in Kenya have emotionally too difficult.
been exposed to sexual comments about them that 2.3.2 Receiving unwanted sexual images
made them feel uncomfortable, including jokes, stories
Twenty percent of the children surveyed said that someone
or comments about their bodies, appearance or sexual
had sent them unwanted sexual images in the past year.
activities. Older children aged 16-17 years were slightly
This experience was more common among older children
more likely to be subjected to these comments (26%
(26%) compared to the youngest age group (14%). There
compared to 16% of 12-13-year-olds). There was no
were no notable differences by gender. When these 200
difference by gender. Among the 212 children who had
children were asked about the last time they were sent
been harassed in this way, more said they were
these unwanted sexual images or videos, over half said
harassed online – via social media and/or an online
they were targeted on social media. Fifty-nine per cent
game – than in person (see page 61). Children aged 12-13
of children ages 14-15 and 16-17 received these unwanted
were more likely to be sexually harassed in person (40%
images via social media, compared to 41% of 12-13-year-
compared to 30% of 16-17-year olds). And less likely than
olds. The platforms most commonly mentioned by those
the oldest children to be harassed on social media (24%
111 children targeted on social media were WhatsApp,
and 44% respectively). Boys were more likely than girls
Facebook or Facebook Messenger and YouTube, followed
to be sexually harassed on social media (42% and 34%).
by Instagram (12%).
Among the 80 children who said they were last
Children were most likely to receive unwanted sexual
harassed on social media, the most common platforms
content from someone unknown to them, followed
cited were WhatsApp and Facebook or Facebook
by an adult friend or acquaintance, a romantic partner
Messenger. Although Snapchat was only mentioned
(or ex-), and a friend or acquaintance younger than 18.
by 6% of this subgroup of children, it was one of the
The easily-abused anonymity provided by the internet
platforms on which children aged 12-13 were more likely
could help to explain why unwanted sexual images are
to be exposed to sexual harassment compared to older
generally sent via social media and why the offender is
respondents.
someone unknown to the child in one-third of cases. The
The most common offenders of verbal sexual oldest children were most likely to say they most recently
harassment of children were a current or former received unwanted sexual images from a stranger (12-13:
romantic partner, a friend or acquaintance younger 15%; 14-15: 35%; 16-17: 39%).
YES 21%
IN THE PAST YEAR
SOMEONE MADE SEXUAL COMMENTS ABOUT

n= 1,014 children
ME THAT MADE ME FEEL UNCOMFORTABLE

How did you feel?* Who did it?*†
Annoyed A romantic partner (or ex-)

24%
20% 22%
20%
26% 18% A family member
11%
Prefer not to say
It didn't
Embarrassed affect me 10%

26%
n= 212 internet-using children aged 12-17 who were subjected to verbal sexual harassment in the past year.

No one
35%
38% 32%
8% 11% Friend
33%
Some other way
TOP 3
Social media
In an online
12%
In person
game
Sibling
BOTTOM 3
12-17 who were subjected to verbal

sexual harassment in the past year. 2% 1% 1%

On which platform did this happen?*† subjected to verbal sexual harassment in the past year.

Facebook or
WhatsApp Facebook Messenger Instagram
I felt I did not know I did not think I was worried

embarrassed whom to tell anyone would I would get in
trouble
57%
believe me
46% 13%
30% 27% 12% 12%
n= 80 internet-using children aged 12-17 who were most recently

subjected to verbal sexual harassment via social media.
n= 74 internet-using children aged 12-17 who did not tell anyone
the last time they were subjected to verbal sexual harassment.
Disrupting
† Harm
YES 20%
IN THE PAST YEAR
SOMEONE SENT ME SEXUAL

n= 1,014 children
IMAGES I DID NOT WANT

How did you feel?* Who did it?*†
Annoyed A romantic partner (or ex-)

19%
16% 19%
23%
16% 18%
A family member
11%
Embarrassed It didn't
affect me Someone else
1%
Prefer not to say
9%

34%
n= 200 internet-using children aged 12-17 who received unwanted sexual images in the past year.

No one
27%
56% 20%
11% 10% Friend
38%
TOP 3
Some other way
10%
Social media
In an online
In person
Male caregiver
game
BOTTOM 3
n= 200 internet-using children aged 2% 2% 0%

12-17 who received unwanted sexual
images in the past year.
n= 200 internet-using children aged 12-17 who received

unwanted sexual images in the past year.
On which platform did this happen?*†

Facebook or
WhatsApp Facebook Messenger YouTube
I felt I did not know I did not think it
embarrassed whom to tell was serious
enough
55% 49% 15%
30% 26% 17%

received unwanted sexual images via social media.
n= 54 internet-using children aged 12-17 who did not tell anyone the
*These figures represent the most common responses selected by children. last time they received unwanted sexual images.
62 †
Multiple choice question Disrupting Harm in Kenya – Evidence on online child sexual exploitation andHarm
Source: Disrupting abuse
data
2.3 OTHER EXPERIENCES OF CHILDREN THAT MAY BE LINKED TO OCSEA
As shown on page 62, over one in four children who emotionally difficult to tell. The second most common
received unwanted sexual images did not tell anyone response was that they did not know where to go or
the last time this happened to them and 38% turned to whom to tell, followed by the belief that what happened
a friend for support. Eight children told a teacher. Few to them was not serious enough to report. Some also
respondents turned to formal reporting mechanisms said that they did not think anyone would believe or
such as helplines or the police. Among the 54 children understand them (15%), or were worried they would get
who did not tell anyone, the main reason was that into trouble (11%).
they felt embarrassed or ashamed or that it would be
The Continuum of Online and Offline Child Sexual Exploitation and Abuse
Case Study – Child Trafficking and Defilement98 and networking but none of the phones was
In March 2020, a suspect travelled to Kisumu registered in his name. He claimed that his
where he rented a house. On 4 May 2020, intention of travelling with the minor to Malindi
detectives from the AHTCPU in Nairobi, acting was to sponsor his education because the boy
on intelligence, arrested the 71-year-old male came from a low-income family.
European national and rescued one minor aged
14 years in Nairobi. The suspect led the officers The victim told police that the suspect had
to a guest house in Nairobi where he had been promised to give him 500 Kenyan shillings if he
lodging with the minor since their arrival from allowed him to touch him and keep the matter
Kisumu in April 2020. The suspect travelled by secret. The boy also received a cell phone from
a public vehicle and bodaboda (motorcycles) the suspect, who would buy him airtime and
to manoeuvre through the police road blocks data bundles, which the suspect would then use
mounted during the Covid-19 lockdown of to download “pornographic” materials. The boy
Nairobi City. He was prosecuted in a Nairobi law informed police that he and two other boys who
court, in May 2020 on charges of child trafficking were his friends, had got to know the suspect
and defilement. at a neighbourhood shopping centre in Kisumu
and had frequently visited the suspect’s house
It turned out that the suspect had first come to watch TV and play video games. He told
to Kenya in 2012 and stayed in the Coast police that the suspect used to touch them and
region. With the assistance of INTERPOL, police they would watch “adult things” [pornographic
established that he had been incarcerated for videos] on the phone.
drug trafficking in South Korea in 2013 and
served five years there before he returned to The suspect is currently remanded in a prison
the Kenyan Coast in 2018. He informed police in Nairobi. The case is pending before court. For
that he produced CSAM for commercial trading the other two victims, the case is pending under
via Facebook. He had six local mobile phone investigation.
numbers which he used for financial transactions
The types of sexual exploitation and abuse of However, our findings – including the case study
children presented throughout this chapter presented above – also reveal that creating a
illustrate some of the ways that digital distinction between online and offline violence
technologies can be used to harm children. does not always reflect the reality of children’s
Continued...
98. The term “defilement” is the legal terminology in Kenya for penetrative acts with a child, and while outdated, is also used in common parlance.
“A person who commits an act which causes penetration with a child is guilty of an offence termed defilement” Sexual Offences Act, No. 3 (2006)
(Kenya) 8(1)
2.3 OTHER EXPERIENCES OF CHILDREN THAT MAY BE LINKED TO OCSEA
experiences. For example, children can be asked that make children who experience violence
or coerced to share self-generated sexual images, ‘offline’ more likely to also experience violence
and this can happen online, offline, or in both ‘online’.
spaces. In addition, digital technologies can also
be used as a facilitator of sexual exploitation The Department of Children’s Services and the
and abuse. For example, social media or instant UNICEF Kenya Country Office both indicated that
messaging can be used to convince or coerce responses to OCSEA are and should continue
children to meet offenders in person, leading to be embedded within the broader child
to ‘offline’ child sexual exploitation and abuse. protection framework and not handled in a silo.
The data in this report include OCSEA that takes This means enabling OCSEA victims to benefit
place in the online environment, OCSEA that from the same services that exist for other
takes place offline but is facilitated by digital child victims of violence. In addition, OCSEA
technology, and OCSEA that is committed ‘offline’ should be included within existing coordination
and then repeated by sharing it online. mechanisms for violence against children: “For
the Department of Children’s Services, we look
Interviews with various stakeholders show that at OCSEA from the broader perspective of child
systems are not fully adjusted to this reality, protection, maybe within sexual exploitation.
and that OCSEA is sometimes perceived as a We do not want it to be in a silo” (RA1-KY-08-A);
‘new kind of abuse’ that requires an entirely “OCSEA should be part of the regular work on
different response. However, as one respondent violence against children and the regular child
summarised, “The channel [the internet] is just protection coordination mechanism” (RA1-KY-
the variable; abuse is still the same.” (RA3-KY-10-A) 01-B)
The same respondent explained that “People do
not understand the online environment; they Despite this important consensus, there remains
think it is harmless to discuss certain things online a lack of clarity around the responsibilities of
as opposed to physically touching someone.” various agencies in addressing cases of child
(RA3-KY-10-A) Another interviewee, from the exploitation and abuse with an online element.
International Justice Mission, recalled that “In all Furthermore, there are cases where online abuse
the years we worked in that field, we didn't have requires a specialised response, for example in
an outright case of online abuse standing alone. law enforcement investigations involving the use
It had to have a component of contact or offline of digital forensics. In other instances, a lack of
abuse, so that should just tell you about the clear laws around OCSEA make it difficult for law
landscape in Kenya in terms of just appreciating enforcement to act and for children to obtain
that online abuse even without contact is still justice through courts. A former counsellor at
abuse.” (RA4-KY-05-A-justice) Childline Kenya also recalled that “After finding
out the case is an OCSEA case, we would sit
In line with this, our data clearly show that in a case conference and the office of public
only a small proportion of children experience prosecution would be represented. In regard
exclusively OCSEA; almost all children who had to charging the offence, he would advise we
experienced OCSEA in our sample had also continue pursuing it as truancy or defilement or
experienced an instance of in-person sexual, sexual abuse or early marriage but specifically
physical or emotional abuse in the past year. This OCSEA, no. The explanation we got from public
could indicate that OCSEA is an extension of prosecution is that the case could not be taken to
existing abuse already experienced by the child, court as OCSEA because there is no policy on it.”
or that there are a common set of vulnerabilities (RA4-KY-06-A-justice)
2.4 INSIGHTS ABOUT VICTIMS AND OFFENDERS FROM
KNOWN OCSEA AND CSEA CASES
2.4.1 Victims 2.4.2 Offenders

Beyond household survey data which gives an Again, very little data about the profile of offenders
indication of who the OCSEA victims are, very little was able to be identified. One indication came from
quantitative data was identified by the Disrupting the frontline workers’ survey where respondents
Harm team. One identified source was Child described the most common relationships between
Helpline International data – reported by Childline offenders and children in the OCSEA cases that they
Kenya. These data showed that of the 16 contacts worked with were community members over the
concerning CSAM that they received in 2018, five age of 18, followed by strangers (nationals), other
related to exposure to adult pornography,99 three relatives over 18, parent/step-parent, family friend
to online sexual exploitation of a boy, and the and community member under 18. More than half
remaining eight to online sexual exploitation of of the frontline workers (29 out of 50) said that men
girls. In 2019, 189 contacts for online child sexual were most commonly identified as offenders and
exploitation for Kenya were recorded100 – 100 facilitators of OCSEA in the cases that they were
contacts for girls and 89 for boys. seeing.
There were no data available from law enforcement Facilitators101 were reported most often to be
quantifying victims and offenders that was specific members of the community over 18, followed by
to OCSEA offences in Kenya. While offline CSEA is not strangers (nationals) and other relatives over 18:
the focus of this report, data on offline offending is “I have encountered cases of OCSEA by biological
included here as a reflection of possible unreported father, step father, teacher and rescue home/shelter
or unrecognised OCSEA offending. Other areas of manager and caregiver” (RA3-KY-29-A). These
the report or case studies describe how victims and data from frontline social support workers seem
offenders may meet in person but correspond online, to contradict the commonly held assumptions of
or meet initially online and then meet in person. foreigner offenders of OCSEA.
Data on offline offending may thus provide context
in which online offending may occur, or may already In the AHTCPU data about offline CSEA, individuals
be occurring. in the 18-29 age group made up the largest
proportion (48%) of the offenders for whom age
The AHTCPU had 79 offline CSEA cases in its caseload was stated in data from the national specialist
between 2017-2019. The majority (72%) of offline unit’s caseload (n=79). One offline CSEA offender
CSEA victims for whom age was stated (n=76) were in the reporting period was identified as under 18.
aged 13 or over. In terms of gender, 91% of those for The majority (91%, n=72) of offline CSEA offenders
whom age was stated (n=78) were female. All victims investigated by the AHTCPU were adult family friends
in the national specialist unit’s offline CSEA caseload of their victim(s); a further seven offenders were
were Kenyan nationals. As mentioned above, these family members. Three offenders investigated by the
data represent only those cases investigated by the AHTCPU during the reporting period were foreign
national specialist law enforcement unit, and does not nationals. Just one offender was female.
reflect all CSEA cases recorded by police in Kenya.
99. This categorisation of the data was provided by Child Helpline International. It is assumed that this was purposeful exposure of a child to
pornography by an adult.
100. In 2019, Child Helpline International adopted a new definition of online child sexual exploitation for their data. Prior to this only cases involving
CSAM were captured in their metrics.
101. A definition of ‘facilitator’ was explicitly defined for the survey participants to answer this question as: “individuals or entities whose conduct
(behaviour) facilitates or aids and abets the commission of sexual offence against the child (sometimes referred to as ‘intermediaries’).”
2.5 BARRIERS TO CHILDREN SPEAKING TO ADULTS
ABOUT OCSEA
Children in Kenya broadly feel that they can depend on strong interpersonal
networks. In our household survey, 82% of children said it was ‘fairly’ or ‘very’ true
that there is at least one teacher they can go to if they have a problem. As many as
94% of children agreed or strongly agreed that members of their families will help
them if they have a problem. In spite of this, as seen in the previous chapters, about
one-third of children subjected to OCSEA never tell anyone, and those who do so
are most likely to confide in friends. Only a small proportion of children mention
it to their caregivers and even fewer turn to formal reporting mechanisms like
helplines or the police. This is true of all the forms of OCSEA explored through the
household survey.
2.5.1 Four reasons for not telling survey was not knowing where to go or whom to
Data from our household survey, access to justice tell. This may indicate both hesitation about whom
interviews with children, survey of frontline workers to tell and insufficient familiarity with reporting
and interviews with government duty-bearers all mechanisms including helplines, the police and the
indicate that children in Kenya might not report social media platforms they use.
OCSEA due to: A former officer from ChildLine Kenya commented:
Lack of awareness of OCSEA: Children are not “Our reporting protocols are not clear when it comes
taught what OCSEA is and might not perceive to online abuse in general. The way you know when
OCSEA acts as criminal. According to our findings you are sexually abused you can just call 116, you can
from the household survey, 53% of internet-using report to the children’s officer in the sub-county or
children have received sex education and 33% you are going to report to the police station gender
received information on how to stay safe online. desk, when it comes to matters online, it’s really not
Information about OCSEA could be integrated into clear where and how you’re going to report. That is
these activities, which need to reach all children. a challenge to children and the public.” (RA4-KY-04-
A-justice) (click here to read more about the issue of
The lack of awareness of many children about OCSEA distinguishing between online and offline abuse in
also reflects a wider lack of awareness in society. investigations).
According to the caregiver of a child who has been
through the process of accessing justice: “More Almost all of the survivors we spoke to during access
awareness needs to be created on the online sexual to justice interviews and survivor conversations in
exploitation for children, caregivers and legal officers Kenya said that they only spoke to the police after
because it is real and people are afraid to report the being helped by others to reach that decision.
cases.” (RA4-KY-05-B-caregiver) Fear of repercussions: One frontline worker
Lack of knowledge of reporting mechanisms: One indicated that fears of being denied access to the
of the most common barriers to the reporting of internet and social media, as well as stigmatisation
OCSEA mentioned by the children in our household (see below), might lead children to avoid reporting
abuse: “Factors such as stigmatisation make children experienced something that bothered or upset
shy away from sharing their experiences especially them online.
when it comes to online abuse. Technology is
presumed to be fashionable hence catching up with Shame, stigma and victim-blaming: According
technology is the dream of everyone and therefore it to our household survey of children in Kenya
will be difficult for a child to share [concerns] since aged 12-17, feelings of embarrassment or shame
he/she might fear that he/she will be denied access” stopped them from talking to others about
(RA3-KY-19-A) their OCSEA experiences. They say it would be
emotionally too difficult to speak out, worry
Such fears may be well founded, as 31% of the about getting into trouble, or feel that no
caregivers in our survey said that they would restrict one would believe them or understand their
a child’s internet use if they knew that the child had situation.
Social and cultural barriers to disclosing In other research, women in East and Central
OCSEA in Kenya Africa report that disclosure could reduce their
The fact that many children subjected to OCSEA marriage prospects and result in stigmatisation by
do not tell anyone, particularly an adult, can be family and community members, which could be
attributed in part to taboos and stigma around another reason for not telling anyone about such
sexual experiences. In fact, 31 of the 50 frontline experiences.102
social service providers we surveyed believed
that stigma from the community influences One parent of a child who had reported and
the reporting of OCSEA in Kenya. Likewise, 60% gone through the justice process said that their
believed that taboos around discussing sex child struggled to: “digest the whole defilement 103
and sexuality influence the reporting of OCSEA. ordeal. It was very traumatising at first because
According to one frontline worker: she was blaming herself and because she was
questioning herself and what other people might
“My society finds it challenging to transition say, she was questioning her integrity and she
children into adults through building their didn’t know what the future would look like. She
confidence and esteem to communicate on was afraid that people would laugh at her, so she
matters of sexuality or abuse. A child is expected was I don’t know if I can say apprehensive, she was
to be shy as a sign of respect so speaking out confused in the beginning because of her morality
becomes a concern.” now that she had been defiled. So that was a very
(RA3-KY-04-A) difficult period for her, but she was counselled,
and she has been able to overcome those issues.”
(RA4-KY-10-B-caregiver)
Continued...
102. Boudreau, C. L., Kress, H., Rochat, R. W., & Yount, K. M. (2018). Correlates of disclosure of sexual violence among Kenyan youth. Child Abuse &
Neglect, 79, 164–172.
103. The term ’defilement’ is the legal terminology for sexual assault in Kenya and used in common parlance.
2.5 BARRIERS TO CHILDREN SPEAKING TO ADULTS ABOUT OCSEA
Under-detection and under-reporting of male under the Kenyan Penal Code, 107,108 and this may
child sexual exploitation and abuse is a global affect the protection accorded to children aged
problem, due to a range of social and legal 12-17 years sexually exploited by an offender
implications.104 One reason is that a child abused of the same-sex. These children may fear legal
by an offender of the same-sex may have difficulty consequences if they report the case. When
reporting the offence due to the stigma associated sampling children who had accessed the justice
with homosexuality.105 Norms about masculinity system to interview them for Disrupting Harm, we
and fear of being viewed as a homosexual have were unable to identify a single male child victim,
been identified as common reasons for the non- even though our survey data clearly shows that
disclosure of sexual abuse by boys in Kenya.106 boys experience OCSEA to a similar degree as girls.
Moreover, “unnatural intercourse” remains a crime This suggests that boys may not feel safe to report.
2.5.2 Why children subjected to OCSEA • As time passed, it became difficult to say anything:
hesitated to disclose “I didn’t tell her because I felt she would be so hurt
The stories of children subjected to OCSEA captured and blame me and reprimand me for not reporting
in our survivor conversations in Kenya graphically it earlier.” (RA5-KY-02-A)
illustrate how the above factors combine to block or • Some of the children also expressed a fear
delay the reporting of OCSEA offences. that disclosure would cause a rupture in their
relationships: “I would just tell my mum and
• None of the nine Kenyan girls who we talked to in she would fix it for me but when this incident
our survivor conversations reported the requests they happened, I didn’t want to embarrass her. She is
received for sexual chat or face-to-face meetings to my mother and she has raised me and provided
the social media platforms they were using – perhaps everything I needed. I didn’t want to make her feel
partly because they did not grasp the malign intent bad, why would I tell her and spoil everything we
of the person they were talking to: “He always said had?” (RA5-KY-01-A)
nice things to me and I ended up trusting him so
• There was also an explicit need for the children to
much and I thought he wouldn’t harm me, and that’s
protect themselves from others. As one of them
why I am still wondering why he would do this to
explained: “It was kind of hard because I didn’t
me and after the incident, I couldn’t find him online
want people to judge me because nowadays not
anymore, maybe he had blocked me and we never
everyone is willing to help; they will just listen to
spoke again.” (RA5-KY-01-A)
your story and spread it all over and since I was in
• The girls also reported feeling ashamed and that high school brutal because you can trust a friend
they had let themselves and others down: “I did and you know this one is a friend and then after
not go straight home because I did not know listening they would go and spread it so I just
how I could explain that to my mother, since she felt it was better to keep it to myself...” (RA5-KY-
had warned me against using phones... That I was 02-A) However, this young person did confide
underage.” (RA5-KY-09-A) in her teacher’s daughter, who then shared the
information with the teacher.
104. Josenhans, V., Kavenagh, M., Smith, S., & Wekerle, C. (2019). Gender, rights and responsibilities: The need for a global analysis of the sexual
exploitation of boys. Child Abuse & Neglect, 110, 4.
105. Ibid., 6.
106. Boudreau, C. L., Kress, H., Rochat, R. W., & Yount, K. M. (2018). Correlates of disclosure of sexual violence among Kenyan youth. Child Abuse &
Neglect, 79, 164–172.
107. Republic of Kenya. (1930). The Penal Code of Kenya (Cap. 63) (Rev. 2012). Section 162.
108. Human Rights Watch. (2019). Kenya: Court Upholds Archaic Anti-Homosexuality Laws.
• When disclosures were made about the abuse
it was often to organisations outside the family:
“Here I have written got help from someone... When sampling children
from Love and Hope... it gave me a new beginning who had accessed the justice
and a chance to open up...” (RA5-KY-01-A) – or to
someone known within the community: “So I went system to interview them for
to a counsellor in our neighbourhood who advises Disrupting Harm, we were
girls.” (RA5-KY-09-A)
• Of the nine children who took part in the survivor unable to identify a single
conversations, four became pregnant as a result male child victim, even though
of the sexual assault, which forced the disclosure
of at least some of what had happened. Mothers our survey data clearly shows
in particular were seen as supportive: “When I that boys experience OCSEA to
realised I was pregnant I told my mother. She did
not judge me; she decided to stand by me. My a similar degree as girls.
parent surprisingly accepted me.” (RA5-KY-07-A)
This included providing care for the baby: “… And
when I was pregnant I told her about it and she
convinced me to keep it and assured me I could
count on her and that was a huge weight off my
shoulder. I cried because I didn’t think she would
react like that.” (RA5-KY-04-A)
3. RESPONDING TO
ONLINE CHILD SEXUAL
EXPLOITATION AND
ABUSE IN KENYA
This chapter presents evidence about current Kenyan response mechanisms. This
includes formal reporting options, and responses by police and the court system.
Finally, it considers the contributions which government, civil society and the
internet and technology industry make to combating OCSEA in Kenya. Much of
the data is drawn from qualitative interviews with government, law enforcement,
court professionals and children and caregivers who accessed the formal justice
system. Responses may not reflect the full range of experiences of those accessing
the Kenyan response mechanisms to OCSEA.
3.1 FORMAL REPORTING MECHANISMS
As seen in the previous chapter, few children report cases of OCSEA to formal
reporting mechanisms like the police or helplines.
In our household survey of 1,014 caregivers, 63% The main channels through which children and
said that they would tell the police if their child was adults can report cases of OCSEA are the Childline
subjected to sexual harassment, abuse or exploitation. Kenya 116 helpline, civil society organisations,
Others said they would not report the incident due the police, and the online reporting portal of the
to concerns about negative consequences, fear of not National Kenya Computer Incident Response Team
being treated properly and/or the belief that reporting Coordination Centre, which aims to counter internet
would have no effect. crime including CSAM offences.
In our survey of 50 frontline workers, 29 said that Child hotlines and helplines
they believed OCSEA cases are not being reported There are several channels through which children
because services are not trusted. and adults can report cases of OCSEA. These include
child hotlines and the one child helpline. OCSEA
The ability to recognise OCSEA and knowledge about
hotlines focus on working with industry and law
how to report may also affect the level of reporting.
enforcement agencies to take down content, they
“Currently, there is little awareness among the public
often now use web-based formats rather than phone
on OCSEA and how to report it” said the head of the
numbers. While some child helplines may specifically
Children Division and Anti-Female Genital Mutilation
focus on online child sexual exploitation and
Unit of the Office of the Director of Public Prosecutions,
abuse, they generally tend to respond to a broader
in one of our duty-bearer interviews. “There is therefore
range of child protection concerns. Some might
a need to sensitise the public so they can be reporting
provide immediate crisis support, some referral
these emerging cases.” (RA1-KY-07-A)
and sometimes ongoing counselling and case
management services.
3.1.1 Childline Kenya 116 and civil society 3.1.2 Law enforcement
organisations Reports of OCSEA can be made at any police station.
Childline Kenya was founded as an NGO by Plan In addition, Kenya has the specialised AHTCP unit in
International, SOS Children’s Villages and the Nairobi, mandated to respond to OCSEA cases.111
Kenya Alliance for the Advancement of Children. A second Unit has also been established in Mombasa
Established in 2006, it is the only dedicated 24/7 and a third is planned for Kisumu.112 These Units are a
toll-free phone helpline, online chat and SMS product of close partnership efforts between Kenya
service for children to report abuse and other issues and other international law enforcement agencies. A
of concern. Childline Kenya is a member of Child total of 27 police officers serve in the two existing units
Helpline International and operates in partnership but only a total of five are dedicated to investigating
with the Department of Children’s Services, in OCSEA cases.
close cooperation with the Directorate of Criminal
According to interviews with representatives from
Investigations and AHTCPU.
the AHTCPU (RA8-KY-01), OCSEA cases are reported
Duty-bearers and frontline workers interviewed or to them by:
surveyed for Disrupting Harm considered Childline
• members of the public
Kenya 116 to be a critical stakeholder in the response
to OCSEA. The helpline did not receive any reports in • civil society organisations
2017. However there has been a big increase between • children’s officers from the Department of
2018 and 2019 (16 and 189 cases, respectively).109 In Children’s Services
2019, OCSEA cases accounted for 78% of reported • the National Kenya Computer Incident Response
CSEA. In comparison, OCSEA cases made up only 2% Team Coordination Centre
of total contacts received by the hotline that year.110 • the Directorate of Criminal Investigation
Eighty-six per cent of calls to Childline Kenya were cybercrime unit
information enquiries related to issues such as child
• regular police stations
maintenance.
• NCMEC
A representative of the Communications Authority • the INTERPOL National Central Bureau.
of Kenya indicated that Childline Kenya lacks
financial and human resources to adequately National Kenya Computer Incident Response
respond to OCSEA: “Concerning support of child Team Coordination Centre
victims, the Child Helpline does this but they are The National Kenya Computer Incident Response
severely incapacitated and don’t have sufficient Team (KE-CIRT) is the core agency mandated to
personnel to tackle this issue [OCSEA] which is remove or take down harmful content reported to
time-consuming. The work they do is commendable, it by the general public and public institutions and
but the government needs to offer the strategic agencies. Its coordination centre has an online Child-
commitment that would strengthen their role Related Cyber Incident Reporting system,113 an email
both from a reporting perspective and from the address and two dedicated hotline numbers for
perspective of providing psychosocial support to reporting internet material suspected of being illegal.
victims.” (RA1-KY-03-A)
109. 2017 data submission confirmed by Child Helpline International, November 2020.
110. These percentages have been calculated using statistics provided by Childline Kenya and Child Helpline International. For 2018, cases of ’child
prostitution‘ and ’sexual abuse‘ were combined to calculate the number of CSEA cases. For 2019, cases of ’sexual abuse’, ’child prostitution’, ’sexual
abuse (sodomy)‘ and ’sexual abuse (incest)’ were taken into account. Other categories that might be relevant in some cases (such as child marriage or
female genital mutilation) were consciously excluded to provide the closest possible figures for explicit (rather than potential) offline CSEA.
111. See: National Police Service Directorate of Criminal Investigations: Anti-Human Trafficking and Child Protection Unit.
112. Both AHTCPU units are well equipped with relevant tools for investigation of OCSEA, including Celebrite, mobile forensic laboratories and internet
connections. While the AHTCPU in Nairobi conducts preliminary investigation through triage of seized OCSEA gadgets using Celebrite for the rapid
identification of victims and evidence to assist in the immediate judicial outcome, the unit has no certified forensic experts and relies on expertise
from the Cybercrime Unit at the DCI Headquarters for detailed analysis and written expert’s opinions for purposes of prosecution (RA8, INTERPOL).
Further, Kenya lacks a national image database on OCSEA. This is understood to have been discussed by the Technical Working Group on Child Online
Protection (see Chapter 3). The existing Child Protection Information Management System under the Department of Children’s Services does not
currently capture data on OCSEA.
113. See: Child Related Cyber Incident Reporting Form.
This Team does not undertake criminal investigations • The Principal Children’s Officer from the
of CSAM cases itself but refers them to the AHTCPU. Department of Children’s Services added: “If you
In our duty-bearer interviews, a Team official want to report to the AHTCPU, you have to call an
explained: “When we receive cases, the initial individual you know within that unit, and tell them
investigation – triaging – is done by the Kenya there is this case, then that person passes it on. It
Computer Incident Response Team to determine should not be like that. Suppose you don’t know
what direction will serve the best interest of the any individual who is working there, does that
child. If it is counselling, that is a decision that can be mean the case will go unreported?” (RA1-KY-08-A)
made and we direct the case to a pool of counsellors;
if it’s the criminal investigations, we will liaise with The Head of the AHTCPU confirmed that reporting
our sister agency which is the AHTCPU to run with has been a challenge: “Reporting is something that
the investigations. We continue providing technical we have been working on as it has been a challenge.
support for them to finalise the investigations.” (RA1- We have got an email address, and then we use the
KY-11-A) Directorate of Criminal Investigations as well as 116.
But there is no number.” (RA4-KY-08-B-justice)
According to one professional working with children,
“The portal for the Communication Authority of Despite these problems, a representative of the
Kenya is not very user friendly. I have used it a couple Watoto Watch Network gave an example of a
of times. It takes you round and round and round. successful referral: “There was a case where a man
If it’s urgent stuff, it’s very difficult for you to report.” was having sexual intercourse with a minor and he
(RA4-KY-02-A-justice) After the Disrupting Harm was recording it. He recorded it and posted it on
interviews were conducted, the system, seems to Facebook so there was public outcry. We got the
have undergone some improvements. For example, report and we did our diligence and reported to the
a ‘Child-Related Cyber Incident Reporting Form’ has AHTCPU and it was taken down immediately and
been added on the reporting portal. that particular man was actually arrested. This is an
example of a case we reported, action was taken
Quantitative data on the number of reports to the immediately and we got to know the person was
KE-CIRT were not available at the time of writing. arrested and the case is in court.” (RA4-KY-03-A-
justice)
3.1.3 Views on reporting to law enforcement
agencies One of the justice professionals interviewed for
The Directorate of Criminal Investigations website 114 Disrupting Harm said that civil society organisations
states that reports can be made directly to the working on matters of child online protection have
AHTCPU through the local offices of the Directorate good working relationships with law enforcement
of Criminal Investigation. However, even public agencies. This working relationship facilitates
officials and professionals acknowledge that flagging of OCSEA cases both for investigation and
reporting to the Unit is not straightforward: for the quick take-down of CSAM. (RA4-KY-02-A-
justice)
• “I would say there is a challenge on how to reach
out to this particular agency. We need a number.
I cannot just be raising awareness and saying, ‘Go
talk to the AHTCPU’. Sometimes it’s not so simple
if you are in Mandera [a county in Kenya], you
Reporting is something that
know, you need a way. That’s a lapse in terms of we have been working on as it
reporting,” said an interviewee from the justice
sector. (RA4-KY-03-A) has been a challenge.
114. See: National Police Service Directorate of Criminal Investigations. (The website also provides links to Childline Kenya and the police toll-free
hotline 112).
Moreover, thirty-five of the 50 frontline workers

whom we surveyed indicated that at least one case
of OCSEA which they had managed directly within The new dedicated online
the past one year resulted in a complaint filed to the portal for reporting OCSEA,
local police/judicial authorities.
launched in 2021 by the
Nevertheless, there appears to be a need to clarify
procedures for reporting directly to the AHTCPU and AHTCPU, will improve
to make it more accessible for organisations working Kenya’s OCSEA investigation
on OCSEA and the general public. In addition, civil
society organisations and the general public may collaboration through
not be well informed about the roles of different connections to other hotlines
agencies. In the words of a justice professional
dealing with OCSEA cases, “I think there needs to be via the INHOPE network.
clarity between the two agencies – the AHTCPU and
the National Computer Incident Response Team.
They are both government agencies but we need to
understand exactly what we need to tell the public
during awareness raising sessions about the category portal for law enforcement authorities in Kenya to
of cases to report to which agency. Right now, it’s report cases of OCSEA.115 Internet Watch Foundation
confusing.” (RA4-KY-03-A-justice) analysts in the United Kingdom will assess reported
CSAM and block and remove it from the internet if
Our interviews with representatives of civil society necessary.
organisations that support the reporting of OCSEA
cases point to a need to establish appropriate At the regional level, KE-CIRT collaborates with
mechanisms for sharing evidence. Currently, some its East African peers under the East African
civil society organisations share evidence (CSAM Communications Organization Cybersecurity
images/videos) with law enforcement authorities Working Group, which is chaired by Kenya.116 On
through informal mechanisms such as WhatsApp the global level, it works with the International
groups. While done with good intentions, this is Telecommunication Union and various national
illegal as it contributes to the circulation of CSAM. computer incident response teams.
3.1.4 International reporting KE-CIRT does not focus solely on OCSEA, but
As seen in Chapter 2, NCMEC received 16,108 reports addresses all kinds of cyber-crimes. This may affect
for Kenya in 2018 and 12,788 in 2019. The vast majority its cooperation with hotlines around the world. The
are reports from technology companies. The Nairobi new dedicated online portal for reporting OCSEA,
office of the AHTCPU is responsible for receiving launched in 2021 by the AHTCPU, will improve
these reports, which are disseminated daily. Kenya’s OCSEA investigation collaboration through
connections to other hotlines via the INHOPE
On 27 January 2021, the AHTCPU and the Internet network. It will be important for KE-CIRT and the
Watch Foundation launched a dedicated online AHTCPU to work together in this context.
115. Internet Watch Foundation. (2021). Kenyan reporting portal to play ‘huge role’ in improving global internet safety.
116. See: National KE-CIRT/CC: Partners.
3.2 LAW ENFORCEMENT RESPONSE
3.2.1 The law enforcers As the Principal Children’s Officer from the
The AHTCPU and regular police stations are Department of Children’s Services explained: “The
mandated to investigate the OCSEA cases reported AHTCPU is a victim-centred unit, so whenever an
to them. Regular police units may refer cases of OCSEA case is reported, we [the Department of
OCSEA to the specialist unit, but they are not obliged Children’s Services and the Anti-Trafficking Unit] sit
to do so. Moreover, many police officers may not be down together, to know where the location of this
aware of the unit’s existence. On the other hand, the case is. If it’s somewhere nearby we go there, if it’s
AHTCPU has the authority to take over OCSEA cases somewhere outside of Nairobi, then I get in touch
from regular police stations for further investigation. with our officers there, then the investigators from
the unit travel there for investigations purposes.
AHTCPU: Duty-bearers and justice professionals When we get to our officers on the ground, if
interviewed for Disrupting Harm regard the it’s a case of rescue, we rescue even before the
establishment of this unit as a great success – not investigators go down there, because the safety of
least because it has catalysed action in other the child is paramount, and we start providing the
mandated government agencies, including the support that is needed by the child. We also try to
Department of Children’s Services and the Office of get in touch with the family, because it’s not just
the Director of Public Prosecutions, which now have about the investigation, it’s the whole spectrum of
officers dedicated and trained in handling OCSEA psychosocial support.” (RA1-KY-08-A)
related cases.
International Cooperation on Law Enforcement collaboration.” (RA4-KY-05-A-justice)

• Through the AHTCPU, Kenya has become the • Kenyan law enforcement has also been
first country in East Africa to be connected cooperating with embassy law enforcement
directly to NCMEC’s CyberTip line, and liaison officers to arrest offenders: “there was a
INTERPOL’s International Child Sexual case of a German who was arrested the other
Exploitation Database. At the time of writing, day. We had already heard that he had been
however, Kenya is not actively using the database abusing children in Kwale. The German Embassy
due to the termination of the internet service informed us that the man had applied for a
provision in the Unit. passport and he would definitely be going for it
• According to a professional from the International (…) When he went to collect his passport, he was
Justice Mission Kenya field office: “For detection arrested.” (RA1-KY-08-A)
and reporting to improve then we must • Several international organisations including
collaborate with international law enforcement UNODC and UNICEF, the United States
like now what we are doing with INTERPOL. For Department of Homeland Security, the UK
example, how we worked in the Simon Harris National Crime Agency and the British High
case. If it were not for UK law enforcement, Commission in Kenya have provided Kenya with
we would have never achieved convictions for mentorship, training, and equipment including
these cases as we didn’t have the pictures, but the forensic mobile laboratory, vehicles,
the pictures were found in the UK and we were computers and desks. (RA8-KY-01)
only able to identify the abuse through that
In addition to its role in investigating OCSEA cases, Protection Unit in my country are doing quite a lot in
the Unit can advise regular police stations and regard to OCSEA and they have been in the frontline
provides training to professionals on handling cases creating awareness and responding to issues of
of OCSEA. In the words of one frontline worker, OCSEA.” (RA3-KY-17-A)
“The Directorate of Criminal Investigations Child
Regular police stations: There are currently five

officers dedicated to the investigation of OCSEA
cases in Nairobi’s AHTCPU. Accordingly, most “[The police officers were
cases are likely handled by regular police stations. asking], ‘You Madam, charge
Some regular police stations have specialised child
protection units with trained personnel, but this is sheet itakaa namna gani?’
not the norm.
(Madam, what will the charge
Our interviews highlighted several challenges for sheet look like?) (...) What is the
investigations by regular police stations:
area of law?”
• Police officers might lack awareness of OCSEA.
A respondent from the Department of Children’s
Services noted that “A person or a guardian may
go to a children’s officer, and what they report as
the presenting case, unless you probe deeper, you what will the charge sheet look like?) (...) What is the
might not know that is a case of OCSEA. (…) We area of law?” (RA4-KY-01-A-justice) In contrast, the
therefore need to empower our officers, our staff, AHTCPU works closely with the prosecutor’s office
even the police, everyone at the grassroots how to charge OCSEA cases correctly, using the available
can they identify a case of OCSEA.” (RA1-KY-08-A) laws. As a representative from the Department
“People do not consider digital crime as a crime of Children’s Services said in our duty-bearer
like physical crimes, so if you say a child has been interviews: “At the [AHTCPU] we have a designated
abused online, you find a policeman will not see prosecutor, so when an OCSEA case is reported, the
the damage,” said the Chief Executive Officer of the police investigators and the prosecutor will sit and
Kenya Film and Classification Board. (RA1-KY-06-A) discuss the case. It makes it so easy as opposed to
Seventy-eight percent of the frontline workers when the police come up with whatever charges
surveyed rated the awareness of law enforcement and we are told this is not how the charge sheet
officers as fair or poor, and 84% their response to should read. So, there is a prosecutor who drafts the
OCSEA, as ‘poor’ or ‘fair’. “Generally awareness of charges so that we avoid the back and forth of the
OCSEA as an important issue of concern is low court.” (RA4-KY-11-A-justice).
among people including law enforcers,” (RA3-KY- • Police officers might lack technical knowledge
03-A) commented one of the frontline workers. on how to handle cases of OCSEA. An AHTCPU
• Police officers may be insufficiently aware of the investigator said, “You find that police stations try
law on OCSEA. According to one Court of Appeals to handle an OCSEA case outside maybe without
judge, “We are not short of laws [but] we have a knowledge on how to seek support or how to do
very serious knowledge gap (…) OCSEA, because the referral. So, then we chip in when they are
of its insidious nature, may not be obvious to a already handling the matter (…) Sometimes we find
local police officer, an investigator, a prosecutor, a they have already done the interrogation of the
magistrate, and even to the victims themselves.” victim without counselling and dismissed the case
(RA1-KY-04-A) A professional with experience in because they are not specialised, so we have to get
OCSEA cases suggested that police officers hesitate a counsellor and we then do the interview process
to institute criminal proceedings in cases where again with different results.” (RA4-KY-08-A-justice)
there is no element of in-person, contact abuse • Police officers trained in handling OCSEA cases
(RA4-KY-05-A-justice). The fact that some forms of may be transferred to other units. One of the
OCSEA are not explicitly covered by law (click here justice professionals interviewed drew attention to
to read an overview of legislation and policy) may the rapid turnover of staff at the child protection
also be a factor here. A former legal officer from units: “When you’re carrying out a training to
Childline Kenya told us some police are not clear police officers, you pick the police officer manning
on how to categorise this: “so they were asking, ‘You the children and gender desk but (…) they keep
Madam, charge sheet itakaa namna gani?’ (Madam, transferring and rotating, so you could have trained
this person, they are so good at their work because (RA4-KY-01-A-child) “I was really free, because they
they understand OCSEA and the unique things did not harass me, they talked to me like anybody
that come along when collecting evidence and else.” (RA4-KY-10-A-child) However, some children
recording the statement of such a case. But the struggled to understand what was said: “At that time,
police service in its normal course of work decided I didn’t know Kiswahili very well and they used to
to transfer the police officer to another unrelated use Kiswahili which I could only understand a bit.”
unit so you are left with someone who is not (RA4-KY-05-A-child) Another child couldn’t keep up
trained.” (RA4-KY-04-A-justice) Another justice with the mixed use of English and Kiswahili stating
professional recommended “that the officers that “as a result I was not confident to ask questions.”
handling children’s cases, the professionals who (RA4-KY-09-A-child)
have been trained on children’s rights and child
protection, these people should not be transferred, Among caregivers, some felt optimistic about the
and if it’s to be transferred, the person brought in involvement of the police: “I was very happy to
their place should be someone handling children’s get the police involved because I want justice for
matters.” (RA4-KY-06-A-justice) my child.” (RA4-KY-01-B-caregiver; RA4-KY-02-B-
caregiver; RA4-KY-04-B-caregiver and RA4-KY-07-
In summary, as a professional from the International B-caregiver) However, others were apprehensive: “I
Justice Mission Kenya Field Office put it, “Having was scared of facing the police, I did not think the
dedicated professionals, you see like the way we police could help, I was afraid of being arrested as a
have the AHTCPU, this is something that we would caregiver.” (RA4-KY-10-B-caregiver)
want replicated around the country. Because what it
means is that you are well trained, it means that now When asked about their motivations for going to the
you’re well-resourced because we know that you are police, caregivers were largely looking for “help with
there and you’re specifically targeting big cases and the case and the arrest of the perpetrator.” (RA4-KY-
it means that you are effective and efficient. Since 05-B-caregiver)
the inception of the unit, it has become easy to make Information about rights and process: Out of the
referrals of cases when you see OCSEA materials. ten children we spoke with in the access to justice
So just having dedicated professionals whether interviews, only three were informed about the
within the police force, whether within the judiciary, process and their rights by the police. To most of
whether within the children’s department, these are the caregivers, it was not clear what rights they and
the most critical people.” (RA4-KY-05-A-justice) their children had. There was a general feeling that
3.2.2 Step by step: What happens when a child the outcome of the police report depended to a
goes to the police? large extent on the officers they met: if they were
lucky, they would be properly informed about the
STEP 1. Children’s and caregivers’ first encounters
process; otherwise they would receive no information
with the police
whatsoever and reported feeling helpless.
Eight of the ten children we interviewed as part
of the access to justice interviews who reported a Six of the ten caregivers felt the police did not
case of OCSEA to the police only decided to engage explain the process to them and their rights properly.
the police after consulting with other people. Six According to one: “I didn’t know my rights. I was
of the ten said they felt comfortable when going to confused about the law, how the case would go.
the police. Of the four who felt uncomfortable, two I didn’t have a headway, so I had to wait until the
did not clarify why. The other two noted that the police explained to me what I should do next, they
discomfort when going to the police was because didn’t tell me my rights as a caregiver of the child.”
“I was forced to go by my family, it pained me” (RA4- (RA4-KY-10-B-caregiver) Another caregiver said: “I
KY-09-A-child) and “I was uncomfortable telling wasn’t told anything, they just sent me back and forth.
them in front of everyone.” (RA4-KY-03-A-child) Someone told me my rights but they were not around
to guide me through the process. I was very sad and
First encounters with the police were generally
the process was tiring. I was hurt as a parent. In the
positive: “I felt good, they tried to help me write
end, I got help from some organisation.” (RA4-KY-04-B-
the statement so they could help me further”
caregiver)
Those caregivers who did receive information about child. (RA4-KY-10-A-child) “I was sent to a hospital for a
the process and their rights were satisfied with this medical examination before being interviewed by the
guidance: “Someone explained the process to me at police,” stated another. (RA4-KY-04-A-child). A third
the station, I was told of my right to testify and I felt child said “They told me they would help me and asked
good about knowing my rights. I got full information on for my mother’s number and if not, they would take me
how I could find justice for my child and as a result, to court themselves.” (RA4-KY-05-A-child) (click here to
I was able to support my child.” (RA4-KY-01-B-caregiver) read more about support services).
Referral to support services: Kenyan law, via the Victim STEP 2. The interview process
Protection Act stipulates that immediate psychosocial The AHTCPU investigator confirmed that AHTCPU
support to the victim117 and the provision of services staff, but not other officers, are trained to interview
to help them deal with emotional trauma118 must be child victims sensitively: “For the AHTCPU, partners
available to all victims of crimes. support us with training. Therefore, officers at the unit
have received special training on how to work with
An investigator from AHTCPU who took part in the child victims. You find however that the other law
access to justice interviews said that child victims of enforcement officers are many but do not normally
OCSEA are given counselling before their statements receive this training.” (RA4-KY-08-A-justice)
are taken. (RA4-KY-08-A-justice) A professional with
experience in OCSEA cases noted that the AHTCPU Child-friendly measures: Our interviewees stated that
in Nairobi is equipped with a room that provides child protection units in regular police stations have
privacy for the victim during counselling and when adopted measures to make the judicial process easier
taking the statement. (RA4-KY-11-A-justice) According for child victims. However, as such units only exist in a
to the AHTCPU investigator, “The counselling room is few police stations, many victims of OCSEA are unlikely
soundproof and has the recording equipment which to benefit.
we use to record the statement if the victim provides
consent.” (RA4-KY-08-A-justice) The children we spoke to in the justice to access
interviews had mixed experiences of interaction with
The same investigator added: “I would say for the police officers. One of the children said that “they used
cases that the AHTCPU handles, because we have words of encouragement” and promised “it will be
that approach where we bring in the Department over soon.” (RA4-KY-01-A-child) However, other officers
of Children’s Services and the Office of the Director made the children very uncomfortable. “The first thing
of Public Prosecutions, most of the OCSEA victims police officer told me was that I was too young to be
get services especially the psychosocial support. (..) pregnant,” one child recounted. (RA4-KY-09-A-child)
When we get the victim, and we bring them here, and
introduce the children officers for counselling then we Some OCSEA victims complained of officers expressing
interview and record statements.” harsh opinions and judging them: “They blamed me
for my situation saying that I asked for it, so they were
While it is unclear how far the Victim Protection Act asking why I was reporting it” (RA4-KY-04-A-child) As a
guidelines are followed in regular police stations, former counsellor pointed out in one of our interviews
several of the caregivers of children who had accessed with justice professionals, “You see when children are
the justice system whom we interviewed confirmed explaining what happened, they [police officers] don’t
that their children received psychological support/ have ‘that’ language. (….) So sometimes the police officer
counselling during the investigation and perceived is forcing them to use the words that you would expect
this as a crucial element which helped their children from an adult. Even culturally, a child is shy to mention
cope with the situation. (RA4-KY-05-B-caregiver, RA4- certain words. Or they don’t know how to describe. They
KY-02-B-caregiver) only say ‘tabia mbaya’ (bad manners). (…) sometimes the
child is recording a statement and the child is crying
Children also described being offered other support and someone (the police officer) is losing patience as
services: “They told me that they are going to take me they want to finish and handle another case.” (RA4-KY-
to a place where I can be taken care of,” recalled one 04-A-justice)
117. Republic of Kenya. (2014). Victim Protection Act No. 17 of 2014. Section 11(2 (c) (ii).
Caregivers of children who had been through the Other issues: Children and caregivers also voiced
justice system expressed the following criticisms: “The suspicions about corruption: “They should have a
police should be friendlier” (RA4-KY-03-B-caregiver; heart instead of taking bribes and destroying other
RA4-KY-05-B-caregiver and RA4-KY-09-B-caregiver); children’s lives [...]” one child said. (RA4-KY-08-A-child)
“The police should adjust their attitude and attend to A caregiver claimed that “it was very hard to get justice
people better […] so that even the children get courage because sometimes the police can be bribed and
to come out and speak the truth” (RA4-KY-08-B- bought to dismiss the case by the perpetrator.” (RA4-
caregiver), and “they should attend to everyone, rich KY-01-B-caregiver) According to another caregiver, “I’m
or poor.” (RA4-KY-08-B-caregiver) “They should create sorry to say but the police like bribes and I think that
child-friendly spaces at the police stations and courts,” was the influence that derailed our case because the
added one of the caregivers. (RA4-KY-05-B-caregiver) perpetrator’s mother would send them money and I
didn’t and that made them take her side and the case
Opportunity to select a police officer: About half of was dropped.” (RA4-KY-04-B-caregiver)
the OCSEA victims we interviewed who had accessed
the justice system (all girls) were allowed to choose Five of the Kenyan girls we spoke to in our OCSEA
who would be in the room when they made their survivor conversations had been supported by staff
statements. Given the opportunity to select a police from NGOs when approaching the police. None
officer, most of them selected female officers. Only of them felt that their cases had been adequately
two felt that the choice of officer was unimportant. investigated. “We went to the hospital and I was
The children who interacted with female officers were examined, then to the police so he could be arrested
more comfortable about sharing details of their stories. but they kept saying they didn’t get him, and the
One child said: “The officer was very friendly and when longer he was missing the more money we were
she asked me the questions, I answered her” (RA4-KY- spending for his search, so we just gave up because
02-A-child) Other children “did not want men to know” the police couldn’t find him but they kept asking
what had happened to them (RA4-KY-03-A-child) for money,” recounted one survivor (RA5-KY-04-A)
or said “There is no way I could tell a man how I was According to another survivor, “They did not help with
feeling.” (RA4-KY-07-A-child) anything (…) They used to call and ask whether we have
found him and we just said we hadn’t found him, and
Caregiver support: Some of the caregivers interviewed they would say that they would see what they can do.”
felt that they were able to support their children during (RA5-KY-05) There was an overwhelming sense that
the process with the police. One caregiver said: “I these crimes were not taken seriously: “I would also like
went with her to the station to write a statement and that those who do such things to girls be put behind
after we went to the hospital for tests.” (RA4-KY-01-B- bars so that they don’t ruin a girl’s future.” (RA5-KY-
caregiver; RA4-KY-02-B-caregiver) However, another 08-A)
caregiver commented that “The whole process was
private and I was just updated afterwards.” (RA4-KY-07- STEP 3. Passing the case on to the Children’s Court
B-caregiver) After the investigation, a case might be referred to the
Children’s Court. There are six dedicated children’s
According to a third caregiver, “They hid her and I courts in Kenya, two are gazetted while the remaining
got reports that she was sick, I felt really bad and it four are not yet gazetted. In court stations without
wasn’t right, something like that, they should allow the dedicated children’s courts, normal magistrates
parents to see their child and support them.” (RA4-KY- courts are converted short term to hear children’s
07-B-caregiver) matters.“OCSEA cases will go to a Children’s Court. If
This mother’s experience of being separated from it is Nairobi, the OCSEA case will go to the Milimani
her child could be related to a tendency from some Children’s Court”, explained a Court of Appeal judge
practitioners to remove children from their homes who is also Chairperson of the National Council on
after reports of abuse even if no-one in the family is the Administration of Justice Special Task Force on
implicated (despite this not being preferred practice by Children Matters (RA1-KY-04-A). Eight out of the
the government). ten children in our access to justice interviews who
had reported OCSEA to the police saw their cases
proceed to court.
3.3 OBTAINING JUSTICE AND ACCESS TO REMEDIES
The Victim Protection Act makes it the duty of courts, administrative authorities and
other persons with functions under the Act to preserve the dignity of a victim at each
stage of the trial and subsequently.119 The Act further provides that each victim should
be dealt with in accordance with his or her age and level of intellectual development.120
According to the Victim Protection Act, every victim deserves protection from secondary
victimisation in all types of proceedings.121 Every vulnerable witness is entitled to legal
and social services at the expense of the State. 122
3.3.1 Court proceedings Act.124 Although the magistrates hearing children‘s

Technical capacity of justice staff cases are gazetted as children’s magistrates, and
In order to deliver justice, courts need to be more are trained in the special safeguards for children
familiar with OCSEA. According to a Judge of the provided under the law, some child-friendly
Court of Appeal: “I think the level of knowledge on measures are tougher to deliver when not working in
OCSEA is nil. Other than knowing that it is a crime a dedicated children’s court. This situation can also
to exploit a child (…). I know there has been some contribute to a backlog of cases.
training, but very little. Very few judicial officers have Participation in the court procedure
been trained.” (RA1-KY-04-A) Only ten prosecutors According to a resident magistrate of the Makadara
in Nairobi have received specific training to support Law Courts, the hardest part for the child is meeting
OCSEA cases, and none in the other 46 counties. the offender “because then when the child walks
(RA1-KY-07-A) in, they see the perpetrator and first thing they do
Child-friendly courts is they freeze.” (RA4-KY-07-A-justice) Every effort
The Children’s Court Practice Directions provide that should be made to ensure that child victims do
“The facilities within the Children’s Court must be not have to face the offender. If this is unavoidable,
distinctive from the ordinary courts and therefore a witness protection box should be used to allow
must be customised to be child-friendly.” 123 However, the child to give evidence without seeing the
there are no specific guidelines that define child- offender (Boxing the offender rather than the child
friendly standards. As a member of staff of the may be preferable). However, only a few courts are
Makadara Law Courts put it, “The Children Act just equipped with witness protection boxes and justice
says child-friendly, so what is child-friendly for an professionals reported that no courts were equipped
autistic child? What’s friendly for a child that has for children to give evidence by video. Mechanisms
suffered trauma? Is it child-friendly the environment, that are the least stressful for the child should be
is it child-friendly the set up? What is this child adopted whenever possible.
friendliness? It has been left to us to define.” (RA4-KY- For a conviction, the victim needs to testify against
07-A-justice) the offender. This can result in re-traumatisation.
Moreover, only a few places have designated Some prosecutors apply for a child to be declared a
Children’s Courts with designated children’s vulnerable witness in line with the law.125 The resident
magistrates who exclusively handle cases involving magistrate puts it as follows: “So most of the times
children. Elsewhere, these cases are handled by when you realise the child is not in a position to
regular courts on specific days of the week in proceed with the matter (…) the prosecutor will say,
accordance with the provisions of the Children ‘We declare the child vulnerable, let the mother
119. Ibid., Section 4(2)(c).

120. Ibid., Section 4(2)(d).
121. Ibid., Section 4(2)(f).
122. Ibid., Section 4(2)(g).
123. Republic of Kenya (2016). The Children Act (Protection of Fundamental Rights and Freedoms of the Child) Children Court Practice Directions ,
Section 28.
124. Republic of Kenya. (2001). Children Act No. 8 of 2001. Section 74.
125. Republic of Kenya. (2006). The Sexual Offences Act No. 3 of 2006. Section 31(1).
of the child speak on behalf of the child’. Then we
proceed with the mother explaining what the child
is trying to say.” (RA4-KY-07-A-justice) During our “The Children Act just says
interviews with children who had accessed the child-friendly, so what is child-
justice system, we encountered one case where a
children’s officer spoke on behalf of the child, thus friendly for an autistic child?
protecting her from the need to re-tell her ordeal
What’s friendly for a child that
in court. The child still felt included and heard, even
though someone else spoke on her behalf. (RA4-KY- has suffered trauma? Is it child-
09-A-justice)
friendly the environment, is it
At least seven of the respondents in our interviews child-friendly the set up? What
with justice professionals noted that in instances
where child victims of sexual abuse were called is this child friendliness? It has
upon to testify, they always testify in closed court in
been left to us to define.”
accordance with the Children Act.126 They described
this as standard practice: “When listening to a
children’s case, (..) the court is empty. That is normal
practice.” (RA4-KY-04-A-justice) Nevertheless, some
children recounted having had to speak in an open there was another OCSEA case already ongoing in
court. Lamu involving a security guard which is handled
by the Department of Criminal Investigation Cyber-
Although it is not always done, the victim may
Crime Unit 127 and I believe it is still in court.” (RA4-KY-
also be asked to participate in a victim impact
08-A-justice)
assessment to guide the judge in the severity of
the sentence. This could be particularly important According to a civil society justice professional:
in cases concerning OCSEA-related offences “When we report a case of OCSEA, we try to follow up
committed without in-person contact, to increase to see if the child has been rescued or if the case has
the understanding of the impact that online child progressed to court but you know for some reason
sexual exploitation and abuse without in-person online cases take forever. We get to hear maybe six
contact can still have. months later something has happened.” (RA4-KY-02-
A-justice)
Duration of process and trial
Rapid processing of OCSEA cases would allow According to an interviewee from the AHTCPU, some
the child victims to move on quickly and reduce of the delays caused by issues specific to OCSEA
the costs families face as they seek justice for cases – namely, a lack of specialised knowledge
OCSEA victims. However, at least half of the justice and understanding of OCSEA within the judicial
professionals we interviewed cited the time it takes professions, especially outside Nairobi, and delays
for cases to come to court and be heard as an by internet service providers in sharing information:
additional source of re-traumatisation. “Challenges come in when you are involving industry
like ISPs. (…) When we request for IP address data,
An official of the AHTCPU informed us that all of the
you find they take time to respond or they don't even
OCSEA cases that had been filed as of March 2019
respond at all. Because we are relying on that to even
(when the unit was established) are still in the initial
reach the victim and the offender, at times, it is a
stages within the criminal justice system: “The very
challenge.” (RA4-KY-08-A-justice) (See also chapter
first operation to arrest perpetrators of OCSEA was
3.4.3 on cooperation between law enforcers and ISPs
done in March 2019 and all the cases are still in the
and global platforms).
initial stages before the court. Before March of 2019,
127. We were not able to interview the Cyber Crime Unit to get more details of the case.
In spite of the delays mentioned, two of the children Victim-blaming and stigma were also highlighted
we interviewed appreciated the pace at which their by the justice professionals interviewed. The role of
cases moved through the judicial system. (RA4-KY- children can be unfairly emphasised in grooming
07-A-child, RA4-KY-05-A-child) cases. Cases involving male child victims with a same-
sex offender entail additional issues. “They are very
Experiences of children in court defensive about it,” one prosecutor noted, “They don’t
An overwhelming majority of the child victims want to admit. It’s a mixture of guilt, remorse and
accessing the justice system whom we interviewed shame. So, for girls it comes across a bit differently
stated that they would be unwilling to interact than for the boys from what we have seen. (….) the
with criminal justice actors again in the future – boys need more counselling to be able to open
a significant deterrent for victims to seek justice. up.” (RA4-KY-09-A-justice) The fact that “unnatural
Negative experiences were mainly related to: intercourse” is criminalised by Kenya’s Penal Code128
may compound the stigma associated with same-
• A sense of exclusion: One caregiver observed sex relations to which boys who are victims of
that the hardest part for the child was the feeling online grooming by male offenders, for example, are
of helplessness: “My child felt deceived by the exposed, regardless of their sexual orientation.
perpetrator and also did not understand what was
going on [at court].” (RA4-KY-06-B-caregiver) Among one of the better experiences, one of the
caregivers interviewed said their child coped well
• Re-traumatisation: Almost all the children were
as “She was treated well at the court and she was
upset by having to recount their ordeal repeatedly
comfortable, and when we got to Nairobi, we were
to different people including the person that
offered free transportation to the court, which took
helped them make the decision to go to the police,
two days.” (RA4-KY-06-B-caregiver). In addition,
various law enforcement officers, lawyers and finally
two of the children felt their experiences were
the court. They found it hard to talk about their
empowering: “I felt the judge listened to me and I
experience to “strangers” and to “talk in open court –
felt like I was free to ask questions.” (RA4-KY-07-A-
it felt like there was no privacy.” (RA4-KY-04-A-child)
child) Another child appreciated the reassurance she
• Language and communication barriers: The heard from a female judge: “She said I would move
use of English, Kiswahili or a mixture of different to a better place and I would deliver my child safely.”
languages made it hard for some caregivers and (RA4-KY-09-A-child)
child victims to follow proceedings. This was made
worse by the suggestion that officials did not use Legal aid
age-appropriate language. Police, judges, court The Children Act entitles every child to legal
staff, prosecutors and lawyers may not be trained representation at the expense of the State129 through
to use child-friendly/victim-sensitive language and the National Legal Aid Service.130 In practice, however,
procedures. Interpretation facilities may not be of the eight child victims we interviewed whose cases
available in the child’s mother tongue. made it to court, only three had access to a lawyer. One
• Victim-blaming and stigma: One caregiver noted of the reasons for this situation could be a shortage
that some questions asked by the judge made of available lawyers. In two cases, there were no
the child uncomfortable. (RA4-KY-10-B-caregiver) government-appointed lawyers available. Nonetheless,
In contrast, one of the children stated that the the caregivers felt sufficiently supported by social
hardest part of her interaction with criminal justice workers who were present at the court hearings.
workers was that “They did not listen to what I On the other hand, a representative of the National
wanted, they refused to release my perpetrator.” Legal Aid Service told Disrupting Harm researchers
(RA4-KY-09-A-child) This child victim, who was that they rarely receive applications for support from
pregnant as a result of the abuse, still intended to child victims of any kind of crime.131 A follow-up with
pursue an intimate relationship with the offender. at least two justice professionals suggested a lack
128. Republic of Kenya. (1930). The Penal Code of Kenya (Cap. 63) (Rev. 2012). Section 162.
130. Republic of Kenya. (2016). Legal Aid Act No. 6 of 2016. Section 36(1).
131. Information provided by telephone on 11 November 2020.
of awareness about this service. Very few know that
this service it is available to child victims (of OCSEA
and other crimes) as well as child offenders, and the The majority of the frontline
procedures for applying. (RA4-KY-11-A-justice, RA4- workers that we surveyed rated
KY-01-A-justice)
the legal services for OCSEA
According to a representative of the Department of
Children’s Services, “The National Legal Aid Service cases as either ‘poor’ or ‘fair’.
has branches in various parts of the country, but
people don’t make use of them because they
don’t know about them. There is a need to create
awareness and hold legal clinics in communities so
people can know that they exist. People do not know 3.3.2 Compensation
what they do.” (RA4-KY-11-A-justice) Where compensation claims are addressed along
with the criminal case, families do not need to take
In Nairobi, the gap in legal aid for OCSEA cases may
additional action, and children and families do
be of less consequence as cases are prosecuted
not have to re-live their trauma again in a separate
by public prosecutors. The officer-in-charge of
compensation case.
the AHTCPU explained: “For Nairobi, even if we
don’t have lawyers watching brief on behalf of the The law in Kenya provides for victims of OCSEA to
victim of OCSEA, we have not felt a gap in legal be compensated by the offender upon conviction.132
representation. This is because the cases in Nairobi However, applying for compensation under criminal
are prosecuted by prosecutors from the Children’s law is not yet an established practice. Of the OCSEA
Division under the Office of the Director of Public victims and caregivers in our access to justice
Prosecutions. They are also lawyers and hence we interviews, only one child was aware of her right to
don’t feel a gap. In other parts of the country we compensation.
don’t have these prosecutors so there is more need
for legal aid.” (RA4-KY-08-B-justice). When informed of their rights during our interviews,
six of the ten children and six of the ten caregivers
The majority of the frontline workers that we surveyed expressed no interest in pursuing compensation.
rated the legal services for OCSEA cases as either ‘poor’ “Even if they compensated me, they destroyed a life
or ‘fair’, both in terms of their availability and quality. and it can’t be refunded, so you just let some things
One participant mentioned that “Much is focused on go,” said one caregiver. (RA4-KY-03-B-caregiver)
psychosocial support; we still don’t have trained legal
experts to support issues of OCSEA.” (RA3-KY-19-A). Nevertheless, most caregivers agreed it would have
been useful to have known about their right to
Be that as it may, victims with the assistance of a compensation earlier. One child saw compensation
lawyer had a less negative experience in court, and as another way of holding the offender accountable.
those who did not have legal support experienced (RA4-KY-10-A-child)
greater stress: “I went to court alone and did not
understand the proceedings. I was afraid but I had Other barriers to compensation include lack of
to speak, I had no other choice.” (RA4-KY-08-A-child) awareness among justice professionals and the
Two caregivers of children who had accessed the likelihood of the offender being unable to pay.
justice system call for the provision of lawyers “to help
Although the Victim Protection Act establishes a
the family and get legal justice for the family.” (RA4-
Victim’s Protection Trust Fund,133 none of the justice
KY-01-B caregiver, RA4-KY-05-B caregiver) Ideally,
professionals we interviewed knew of any victim
caregivers and child victims should be consulted
that had been compensated by it or provided
before the allocation of a lawyer. There appears to be
with support to attend court hearings. A senior
a preference for female lawyers, at least for female
coordinator of survivor services from the International
OCSEA victims.
132. Republic of Kenya. (2018). The Computer Misuse and Cybercrimes Act No. 5 of 2018. Section 45(1).
133. Republic of Kenya. (2014). Victim Protection Act No. 17 of 2014. Part V.
Justice Mission confirmed that the fund was not yet However, for cases of OCSEA handled at regular
operating, although the board was in place. (RA4-KY- police stations, there is need to create awareness and
05-A-justice). strengthen linkages with the Department of Children’s
Services so that support can be extended to victims.
Out-of-court settlements are undesirable as the
offender is not held accountable before the law. Psychosocial support: Children have the right
There were no such settlements in the cases we to free counselling under the law137 but these
explored through our interviews with child victims and services are not readily accessible everywhere. “It’s
caregivers who had accessed the justice system. One not that counsellors are not there,” explained the
caregiver recalled that “The white perpetrator promised representative of the Department of Children’s
my child that he would give her some money if she Services whom we interviewed, “There are
kept quiet. (…) The perpetrator later refused to give us professional counsellors all over in this country and
the money stating that we had gotten him arrested, you can access them if you are able to pay. So, you
therefore breaking the agreement that if we did not find victims are not able to afford it and for us, we
report the case to the police he would give us money.” work with child focused agencies who can provide
(RA4-KY-03-B-caregiver) for free. But you see we cannot depend on that, as
there are those areas without those child-focused
3.3.3 Social Support Services agencies. If there was a fund that could support
Under the Victim Protection Act, it is the first duty victims to pay for counselling, it would ease this
of a person dealing with a victim to secure the challenge.” (RA4-KY-11-A-justice)
victim from further harm.134 In partnership with civil
society organisations, the Department of Children’s A frontline worker from an organisation undertaking
Services is tasked with providing rescue, shelter and awareness raising and training activities argued that
reintegration services, referrals to medical care135 and “The support and counselling centres are not enough
psychosocial support136 to child victims of crime. and it’s costly. There are no such free services offered
by the government, thus most people don’t take
While children are best protected in a home their children for the same. Limited awareness has
environment, rescue or temporary shelter services also made children and parents not to realise this as
are needed if the situation at home is unsafe or a violation.” (RA3-KY-37-A)
alternative family-based care is not immediately
available. In practice, the availability of support In fact, 68% of the frontline workers we surveyed
services for children recovering from OCSEA is – most of whom provide counselling services
limited. The frontline workers who took part in our themselves – rated the availability of psychological
survey saw location as the main limitation – services services as either ‘poor’ or ‘fair’ and 62% rated the
are concentrated in urban areas – followed by the quality as either ‘poor’ or ‘fair’.
cost and low quality of the services.
Free services could be made more widely available
Referrals: For those OCSEA victims whose cases are by activating the Victim Protection Trust Fund, set up
handled by the AHTCPU, the Department of Children’s under the Victim Protection Act to assist the victims of
Services has appointed a children’s officer to assist the crimes.138
Unit in supporting OCSEA victims. In our interviews
One of the few child-focused organisations
with justice professionals, an official of the Department
providing free counselling is Childline Kenya, which
of Children’s Services and two officials from the Unit
offers children on-location counselling services in
confirmed that whenever a case is reported, the two
Mombasa, Nairobi and Kisumu as well as telephone
agencies work together to ensure that the victim
counselling country wide.
receives the support s/he needs even as the law
enforcement officers undertake their investigations. Medical services: Free medical support is widely

137. Ibid., Sections 11(2) & 14(2).
138. Ibid., Section 28(2)(a).
available. A former legal officer of Childline Kenya
explained that “Mostly when you get cases of abuse,
including OCSEA, you refer them to public facilities Around half of the frontline
for medical attention where treatment is free.” (RA4-
KY-04-A-justice)
workers that we surveyed
thought that medical services
Of the frontline workers we surveyed, around half
thought the availability and quality of medical for OCSEA victims were either
services for children subjected to OCSEA was either
‘good’ or ‘excellent’. This is higher than for any of
‘good’ or ‘excellent’.
the other services the frontline workers were asked
to rate (i.e., psychosocial, legal and reintegration
services).
The representative of the Department of Children’s
Reintegration services: The Children Act recognises Services interviewed for Disrupting Harm confirmed
that a child who has been sexually abused or is that there are government shelters where victims of
prone to sexual abuse and exploitation, including OCSEA may be placed if it is absolutely necessary to
CSAM-related conduct, needs care and protection.139 remove them from their current environments, and
Additionally, The Victim Protection Act outlines the role that partners of the Department offer this service
of the Victim Protection Board in advising the Cabinet in places where there is no government institution.
Secretary on activities aimed at the implementation However, in some counties the shelters are few and
of rehabilitative programmes for victims of crimes.140 far between: “Where we have vast counties, you’ll find
The court before which such a child is brought may children have to travel so many kilometres before you
commit such child to a rehabilitation school suitable to can actually get to a place of safety for them as we
their needs and attainments.141 move them from the perpetrator.” (RA4-KY-11-A-justice)
Protecting children
A number of respondents in Kenya spoke of Fear of removal can also discourage children from
‘rescuing’ children living in particular harmful disclosing abuse and seeking help, since they
circumstances. This is also one of the prescribed may view residential care negatively or even as a
duties of the Department of Children’s Services punishment.142 Some emergency circumstances
under the law. While some emergency still demand removal actions, but where possible,
circumstances demand such action, the removal removing the offender instead can protect the
of a child from their family and community child while maintaining their attachment to their
for protective reasons bears its own risks. organic support systems.
Two other justice professionals interviewed spoke us to say that now they are ready for reintegration.”
of the need to strengthen after-care services. (RA4- (RA4-KY-05-A-justice)
KY-05-A-justice, RA4-KY-11-A-justice) One of them
proposed minimum standards: “There is the issue One child said that professionals who interact with
of holistic survivor care, just having the minimum children should contribute to the reintegration of
standards of care for every child victim of OCSEA, in children back to their homes and their families at
order to know what must be given to every child for the end of their cases: “Stakeholders should follow up
and help all children.” (RA4-KY-06-A-child)
139. Republic of Kenya. (2001). Children Act No. 8 of 2001. Section 119(1)(n).
140. Republic of Kenya. (2014). Victim Protection Act No. 17 of 2014. Section 32.
141. Republic of Kenya. (2001). Children Act No. 8 of 2001. Section 125(2)(c).
142. ECPAT International. (2017). Through the Eyes of the Child: Barriers to access to justice and remedies for child victims of sexual exploitation.
Bangkok, ECPAT International. 54-55.
3.4 COORDINATION AND COLLABORATION
3.4.1 Policy and government between government agencies with mandates

Promising developments and initiatives related to OCSEA, according to most of the
At the policy and government level, our research frontline workers we surveyed. A principal officer
has identified several promising developments and from the National Council of Children’s Services
initiatives in addressing OCSEA in Kenya: described the group as vibrant, well-coordinated
and the embodiment of a multi-sectoral approach,
• Acknowledgement of the threat: In the 12 duty- while at the same time admitting: “All these
bearer interviews, almost all the respondents agencies may not be fully conversant with it and
perceived OCSEA as a growing threat. In the the process and procedures to be followed, and
words of a Principal Children’s Officer from the we are still learning and putting measures in
Child Online Protection Unit of the Department of place to ensure we address such gaps so there
Children’s Services: “As much as online child sexual is seamless flow.” (RA1-KY-02-A) A representative
exploitation and abuse has been there for some of the Communications Authority of Kenya still
time, it was not as it is now. With the advancement felt the need for more synergy: “The agencies
in technology, we find that there is more online know their mandates but (…) for example, when
child sexual exploitation and abuse than what we it comes to reporting of cases and how cases
used to know several years ago. For example, when should be handled, agencies tend to talk about
I was in the field, we would hear of pornographic their mechanisms yet there is need to find ways of
materials, children being put in a room and working together with others.” (RA1-KY-03-A)
somebody trying to come up with some videos but Awareness raising initiatives
that time it was ‘a foreign issue’. It was something • The Ministry of Education is currently integrating
done by ‘Wazungus’ [white people] in secluded child online protection into the new school
and affluent areas. It was not as big as it is now.” curriculum, which will help to standardise
(RA1-KY-08-A) awareness raising in schools. A representative of
• Understanding of the need for collaboration: the Kenya Institute of Curriculum Development
The importance of a multi-stakeholder approach said that OCSEA is not directly addressed but that
for addressing OCSEA was underlined in six of the relevant topics such as online offenders and how to
twelve duty-bearer interviews. According to the avoid them are included. (RA1- KY-10-A)
Principal Children’s Officer cited above, “OCSEA • In partnership with Terre des Hommes
is not just one department’s problem. It cuts Netherlands, the Ministry of Education recently
across government and state agencies” and “For launched the first children’s and facilitators’
the Department of Children’s Services, we look manuals for training on online safety and security.144
at OCSEA from the broader perspective of child These are to be used to equip children and
protection, maybe within sexual exploitation. We caregivers to identify, prevent and respond to
do not want it to be in a silo.” (RA1-KY-08-A) As an abuse and exploitation. It is not clear if they will be
example, the health system also may play a role. rolled out nationally.
For instance, as a gateway to support for victims.
• The Communications Authority of Kenya runs
• Creation of the National Technical Working Group awareness raising initiatives on child online
on Child Online Protection: The group brings protection. Child and adult consumers of
together mandated government agencies, civil information and communication technology
society organisations, industry representatives services can visit the Kikao Kikuu county forums to
and UN agencies.143 Thanks to this working group, discuss and explore solutions for communications
adequate coordination exists at the national level challenges in the counties.145
143. The National Technical Working Group on Child Online Protection is Comprised of: The Department of Children’s Services, Directorate of
Criminal Investigations-Anti Human Trafficking, Child Protection Unit, Directorate of Criminal Investigations HQs – Cyber-Crime Forensic Lab,
Information and Communication Technology Authority, Department of Information, Kenya Institute of Curriculum Development , Judiciary,
Kenya Film Classification Board, Communication Authority of Kenya, Ministry of Education, Kenya Police Service, Office of the Director of Public
Prosecutions, Attorney General’s office, Kenya Computer Incidence Response Team, UN agencies- UNICEF, UNODC, INGOs/ National NGO’s- Terres
de Hommes, Child Line – Kenya, African Institute of Child Studies, Watoto Watch and ECPAT.
144. KBC News. (2020). Education Ministry launches online safety manual.
145. See: Communications Authority of Kenya: Kikao Kikuu.
• The Kenya Film Classification Board runs a media
literacy programme entitled You are what you
consume146 intended for a wide national audience. Where a child is abused, we
Training programmes are not able to reach the child
• In conjunction with Terre des Hommes in good time. We have to refer
Netherlands, the Department of Children’s Services
provided 76 children’s officers, police officers, to partners.
magistrates and prosecutors from Nairobi, Nakuru,
Kisumu and Mombasa with two days of training
on OCSEA prevention and response, since in the
words of the principal children’s officer from the
and that it’s something they need to consider in
Child Online Protection Unit of the Department
their broader work of violence against children.”
of Children’s Services, “It’s a negligible number of
(RA1-KY-01-B)
officers that have been trained on OCSEA.” (RA1-KY-
08-A) • Insufficient training of professionals. Eighty
percent of the frontline workers we surveyed rated
• According to the Ministry of Information
the government’s training efforts as either ‘poor’
Communication Technology, a five-day Child
or ‘fair’. As an example, the Head of the Children
Online Protection Training programme has been
Division and Anti-Female Genital Mutilation Unit
developed by the Communications Authority of
commented: “Awareness and technical capacity
Kenya, African Advanced Level Telecommunications
is a challenge. (…) These cases are technical and
Institute and the International Telecommunications
there are forensic issues as well. (…) Except for
Union Centre of Excellence for professionals from
the few prosecutors trained and working with
both government agencies and non-governmental
the Directorate of Criminal Investigations Child
organisations. Two workshops147 have been
Prosecution Unit, there is need for training not
conducted face-to-face and others online.
only for prosecutors but police and the bench
Challenges [magistrates and judges] as well.” (RA1-KY-07-A)
Nevertheless, much needs to be done. “I do
• Understaffing at the Department of Children’s
acknowledge that conversations have started but we
Services. According to a Deputy Director of the
have a long way to go in terms of awareness creation
Department of Children’s Services, “We have
and setting up of appropriate response structures
officers in 283 out of the 295 sub-counties. In those
and also effective referral pathways,” said one of the
283, you will find in a sub county it is only one
respondents to our frontline workers’ survey. (RA3-KY-
officer without a driver, goes to the post office,
29-A) Pressing issues highlighted from our research
goes to court (….) The Department of Children’s
activities included the following:
Services is under-represented (….) Where a child
• Lack of awareness at local level and among is abused, we are not able to reach the child in
frontline workers. At county level and lower levels, good time. We have to refer to partners.” (RA1-KY-
the existing child protection mechanisms – i.e., the 08-B) The Department’s Online Child Protection
Area Advisory Councils – coordinate OCSEA issues. unit currently has one officer. A representative
However, according to the UNICEF Kenya’s Head of from the Communications Authority of Kenya
Child Protection, “The challenge is that there is still commented: “Robustness on the role and mandate
not much awareness at the county level in regard of Department of Children’s Services needs to be
to OCSEA prevalence. Take a Children’s Officer strengthened. The Department for example on
somewhere in West Pokot, they probably have low the ground works with volunteers which might
levels of awareness that this is an emerging issue not make it a strong approach in carrying out its
mandate.” (RA1-KY-03-A)
146. See: Kenya Film Classification Board: You are what you consume.
147. AFRALTI. (2020). Child online protection training workshop 2020; & AFRALTI. (2019). November child online protection training workshop.
• Limited government funding: Funding is a

challenge for many agencies. An interviewee from
the Department of Children’s services explained “Awareness should be wide
that their work is funded by partners such as including also the communities
UNICEF or civil society organisations: “When it
comes to the financial resources for children’s themselves because if they
services, it’s minimal (…) but when we flag out
our programmes properly, then we can get some
don't report, then there is a
funding from partners, not from mainstream problem. That is where we
government. The [children’s officer heading the
child online protection unit] is funded by partners
need to begin.”
(...) Last year, we did not have any resources from
the government for that work (…) all our activities
were supported by partners, OCSEA being a part
questioned whether awareness raising activities
[of this].” (RA1-KY-08-B) A principal officer from
were reaching all regions and segments of society.
the National Council of Children’s Services added:
“So far I would say they are (…) only reaching the
“We find that the children's sector is inadequately
rich. (…) If there is a programme on TV and you are
funded so that whatever you are given is not
sensitising parents on how to keep their children
adequate (…) We have other government agencies
safe online, there are so many other parents out
with a mandate on children and more often than
there who may not have that TV,” said a Principal
not, in case of any emergency or something, some
Children’s Officer at the National Council of
components are more prioritised – health and
education but they ignore child protection. There Children’s Services. (RA1-KY-02-A) “How many
is, therefore, need for a balance in the distribution people will sit to watch TV?” added a principal
of resources for child protection services.” (RA1- children’s officer from the Child Online Protection
KY-02-A) The chair of the Special Task Force on Unit. “We should look at the medium we use when
Children Matters noted that budgetary resources to we try to sensitise people in the grassroots (…) We
safeguard the rights of children within the criminal should use different approaches.” (RA1-KY-08-A)
justice system is a challenge for the judiciary and The Department of Children’s Services confirmed
other law enforcement agencies alike: “We don’t that awareness activities have been conducted
have any budget that is designated for children. We in a limited number of counties, not including
just use what is given for other court operations. We rural counties. Watoto Watch Network confirmed
know children have their own needs even when that their awareness raising activities under the
they come to court, they require lunch, we require project implemented by UNICEF with funding
to set up the place where they stay, we need a from the Global Partnership to End Violence
caretaker for them, we need toys and games (…). against Children, through its Safe Online initiative,
The police deal with children every day, they have targeted only Mombasa, Machakos, Nairobi and
a child protection unit yet there is no budget for Nakuru. “We need to train more stakeholders
children” (RA1-KY-04-A) In our survey of frontline like ‘training of trainers’ and have them in every
workers, 96% of respondents rated funding as ‘poor’ county, and then now they are able to train others,”
or ‘fair’. Funding and training (rated as ‘poor’ or ‘fair’ remarked a Principal Children’s Officer. (RA1-KY-
by 80%) were most frequently selected as the major 08-A) A representative of the Communications
obstacles to adequate services for victims of OCSEA. Authority of Kenya also acknowledged that “It
• Lack of public awareness: “Awareness is a big needs to go a level deeper in content including
problem because I don’t think we even know there talking about the underlying issues as technology
is a crime like that,” said the Court of Appeal judge is only an enabler and also to broaden the scope.”
who also heads the Special Task Force on Children (RA1-KY-03-A) Of our frontline workers’ sample, two
Matters, “Awareness should be wide including thirds rated the government’s awareness raising
also the communities themselves because if they activities and efforts on “speaking publicly about
don’t report, then there is a problem. That is where child sexual exploitation” as either ‘poor’ or ‘fair’.
we need to begin.” (RA1-KY-04-A) Respondents
3.4.2 Civil society Domestic Internet Service Providers
Civil society organisations in Kenya play a major Evidence gathering: When the law enforcement
part in responding to OCSEA. They refer cases to authorities need evidence from a domestic Internet
the police and the courts and cooperate with the service provider – for example, to identify who was
Department of Children’s Services in the provision of using a particular IP address or phone number at
services like shelter, counselling and legal aid. They the time an offence was committed – they serve
are also involved in awareness raising activities and a court order to the service provider demanding
in training the child protection workforce. “OCSEA this information. They can then use the subscriber
requires a multi-sectoral approach and it’s not for the information to locate and apprehend the suspect
government to work alone without the civil society and to submit as evidence in court.148
organisations,” says a UNICEF child protection officer Article 50 of the Computer Misuse and Cybercrimes
at the UNICEF country office. (RA1-KY-01-A) Act provides that investigating agencies can apply for
The local and international civil society organisations court orders when they have reasonable grounds to
and UN agencies working on OCSEA in Kenya believe that subscriber information in the possession
include Childline Kenya, Watoto Watch Network, or control of a service provider is needed for the
Mtoto News, Terre des Hommes Netherlands, ECPAT purpose of an investigation.149 Moreover, service
providers150 can be compelled to collect, record or
International, Arigatou, UNICEF and the United
cooperate in the collection or recording – by the
Nations Office on Drugs and Crime.
police or other authorised person – of traffic and
While these organisations provide crucial services, content data.151
their efforts and resources alone are not enough.
According to an investigator from the AHTCPU
According to one frontline worker, “The support
interviewed for Disrupting Harm, the Computer
systems for helping survivors of OCSEA are extremely
Misuse and Cybercrimes Act “has helped us to bring
limited due to the small number of NGOs dealing
in OCSEA. This is also due to insufficient awareness on board the internet service providers on what we
by most service providers from the government need to prove OCSEA cases. (…) We can go to internet
on what OCSEA is and how it happens.” (RA3-KY- service providers and they are more compliant
01-A) Another frontline worker commented: “The
because we have the Act as a reference.” (RA4-KY-08-
A-justice)
government must not take for granted the issues
that surround the children especially on the internet. Nevertheless, according to the same source, it
They give less support to organisations that try to is sometimes difficult to obtain the information
deal with online issues. Governments should be requested due to:
supportive.” (RA3-KY-26-A)
• lack of a policy that regulates the length of time for
When asked to assess the collaboration on OCSEA which internet service providers must store data:
among non-government organisations, 52% of “You find that the internet service providers store
frontline workers said it was ‘good’ or ‘excellent’ and data for some time. It is not regulated so at times
48% ‘fair’, ‘poor’ or ‘non-existent’ you don’t get it because they have done away with
3.4.3 Internet service providers and platforms the data as they have their own policy to control
the data, what to store and for how long. If it is data
Collaboration with internet and mobile service providers
that has overstayed, we don’t get it. When they
and platforms is essential to investigate crimes
have it, they comply.” (RA4-KY-08-A-justice)
and prevent the dissemination of CSAM. The legal
requirements and practical procedures differ depending • slow response to the court orders, leading to delays
on whether the operators are Kenyan or global. in prosecutions: “When we request for IP address
148. This approach however ignores challenges posed by carrier grade Network Address Translation, a process by which rapidly exhausting IPv4
addresses have been assigned by ISPs to multiple users at the same time, thereby precluding definitive identification of the device and user behind
an IP address in certain cases.
149. Republic of Kenya. (2018). The Computer Misuse and Cybercrimes Act No. 5 of 2018. Section 50(1)(b) and (2)(b).
150. Ibid., Section 2.
151. Ibid., Section 52(1)(b) & 53(1)(b).
data, you find they take time to respond or they If internet service providers were obliged to report
don’t even respond at all. Because we are relying companies and/or individuals disseminating or
on that to even reach the victim and the offender, distributing CSAM, an investigator from the AHTCU
at times, it is a challenge.” (RA4-KY-08-A-justice) pointed out, “It will assist in that immediately they
detect OCSEA material from their side, it will be
Safaricom, the largest mobile telecommunications controlled and the materials will not be widely
operator in Kenya with a market share of about two- distributed.” (RA4-KY-08-A-justice) However, even if
thirds, has reportedly taken steps to process requests domestic internet service providers were required by
for information for the prosecution of OCSEA law to filter/ block/take down CSAM, this would only
cases faster. The officer-in-charge of the AHTCPU apply to material hosted in Kenya – and the Internet
explained: “Safaricom identified a specific liaison Watch Foundation as well as INHOPE data indicate
officer for this unit. An officer of the Directorate of that relatively little material is currently hosted in
Criminal Investigations who is attached there to deal Kenya.
specifically with the Unit’s requests. We are yet to
have the same with other internet service providers.” In a promising development, in the first quarter
(RA4-KY-08-B-justice) of 2020, the Department of Children’s Services
indicated that it had initiated discussions with
Removing/reporting CSAM: Kenyan laws do not internet service providers to discuss their role in
impose legal duties on internet service providers to addressing OCSEA and other child online protection
filter and/or block and/or take down CSAM and to issues. “They did not even know about these OCSEA
report companies and/or individuals disseminating, cases. When we had that meeting with them and we
trading or distributing the material. However, the educated them, we told them they can innovate and
National Kenya Computer Incident Response Team prevent these abuses even from being uploaded on
Coordination Centre may request internet service their platforms,” explained the principal children’s
providers to remove content from their platforms, officer from the Child Online Protection Unit of the
and compliance with requests is reportedly high. Department of Children’s Services. (RA1-KY-08-A)
As a representative of the Team explained: “If there Global platforms
are gross materials, we in our own volition, can Evidence gathering: If a report is made to the
make a move to request the service providers (…) to Kenyan police about OCSEA on a global platform,
have some content removed. (…) Other government such as Facebook, a request is made to the platform
agencies and also private organisations can flag to obtain subscriber information and IP data. Once
some content and share with us and we will be in a the IP is known, the police then follow the domestic
position to analyse, see if it meets the criteria, and if internet service provider request process to resolve
it does, we will occasion the removal (…) We compel the IP data and confirm the identity, location and
internet service providers to remove when the other details of the suspect.
material in question is of a serious criminal nature –
for example, child sexual abuse [material]. And you The wording of Article 50 of the Computer Misuse
find that there is a way it’s structured, that when it and Cybercrimes Act, which refers broadly to “service
[the notice] reaches their servers, they know this is provider offering […] services in Kenya”,152 opens the
a demand and not a request. A demand is actioned door to obtain court orders directed to non-local
within one hour. We have that arrangement and there service providers. The Act also enables the Office
is no instance where we have had to penalise.” (RA1- of the Attorney General and Department of Justice
KY-11-A) However, the team was unable to provide us to request (or receive requests for) assistance from
with specific statistics on child-related reports. another State in any investigation related to a crime
under the Act.153
152. Ibid., Section 50 (2)(b).

In addition, the Mutual Legal Assistance Act outlines notifications of potential violations of their own terms
procedures for cases where Kenya has to seek legal of service. Since CSAM is contrary to the platforms’
assistance from another state with which it has a terms of service and US law, it would be in the
mutual legal assistance arrangement, or vice versa.154 companies’ interests to remove such content.
Global platforms cannot be compelled to disclose Within this context, the normal procedure would be
information by Kenyan court orders, since they for the National Kenya Computer Incident Response
are governed by the domestic laws in their home Team (KE-CIRT) to send an informal email rather
countries – in the case of the USA, the Stored than a court order demanding removal. KE-CIRT was
Communications Act and Electronic Communication unable to provide statistics on its use of procedures
Privacy Act. US law expressly prohibits the disclosure or the quantity of CSAM removed as a result.
of communications content such as messages
and images directly to non-US law enforcement
authorities.
However, US tech platforms may voluntarily disclose Transparency data

non-content data, which includes subscriber data and In 2017, 2018 and 2019, the transparency reports
IP logs, to such authorities. The largest US technology of major social media platforms show that
companies have dedicated Law Enforcement authorities in Kenya made:
Response Teams (LERT) to respond to data requests
• 14 requests to Facebook for content
from non-US law enforcement agencies. Facebook
restriction, related to violations of hate
now has a dedicated Law Enforcement outreach
speech and election laws, and a private case
manager for Sub-Saharan Africa.
of defamation;
Interviews held as part of the analysis of non-law • 20 requests to Facebook for user data;
enforcement data for Disrupting Harm indicated that
• 19 requests to Google for content removal,
there are informal working arrangements between
mostly related to defamation;
the Kenyan law enforcement authorities and
companies like Facebook and Google in relation to • 9 requests for Google user data;
the voluntary disclosure of non-content data. • 5 requests to Apple;
• 1 request to Twitter for user data, and 1 for
If the Kenyan police need to obtain information on
content removal;
content hosted outside of Kenya but not on a US
tech platform (e.g.: on a website), the request would While none of the major platforms publish
rely on the existence of a mutual legal assistance data specific to OCSEA or fully disaggregated
arrangement with the government in question. by the type of crime, the diversity of platforms
addressed suggests that Kenya engages with
Removing/reporting CSAM: With respect to
US technology companies more than some
removing/reporting CSAM, there are rarely any formal
of the other African countries studied for
agreements between national law enforcement
Disrupting Harm.155
agencies and global platforms. The platforms would
prefer to view requests from government partners as
154. Republic of Kenya. (2011). Mutual Legal Assistance Act No. 36 of 2011.
155. Platforms were selected on the bases of high volumes of reports to NCMEC (10,000+), availability of transparency reporting, and known
popularity in Disrupting Harm focus countries. In addition to US-based companies, transparency reports for Line and TikTok were also reviewed.
4. HOW TO DISRUPT
HARM IN KENYA
Disrupting harm from online child sexual exploitation and
abuse requires comprehensive and sustained actions from
us all – families, communities, government duty-bearers, law
enforcement agencies, justice and social support service
professionals, and the technology and communications industry.
While children are part of the solution, the harm caused by
OCSEA obliges adults to act to protect them; we must be careful
not to put too much of the responsibility on children.
A detailed set of actions needed in Kenya are clustered under

five key insights from the Disrupting Harm data and sign-
posted for different stakeholder groups. However, all these
recommendations are interlinked and are most effective if
implemented together.
4. HOW TO DISRUPT HARM IN KENYA
INSIGHT 1 Awareness and education programmes should be

developed and tested through consultations with
children and caregivers, to reflect their perspectives
Internet-using children in Kenya are of online risks and the techniques they use to keep
subjected to OCSEA. According to themselves safe. Key objectives should include:
children who were subjected to OCSEA • Equipping caregivers with the knowledge and
and frontline workers, most offenders skills to foster safe and ongoing communication
are someone the child already knows. with children about their lives online (see Start the
These crimes can happen while children chat156 for an example).
spend time online, or in person but • Challenging social norms and taboos that limit
discussion about sex and deter children and adults
involving technology. from seeking help about child sexual exploitation
and abuse because of embarrassment and shame.
Government
• Supporting caregivers, many of whom have never
1.1. Adapt and deliver national-scale awareness used the internet, in going online and becoming
and education programmes about the sexual more familiar with the platforms that children are
exploitation and abuse of children – including how using (see Be Connected 157 for an example).
technology might play a role. These programmes
• Strengthening children’s digital literacy to
must be evidence-based and not shy away from
provide them with the skills and understanding
difficult and sensitive messages about sex, or the
needed to avoid or navigate dangerous situations
finding that offenders are often people known to
online. This could include lessons about how to
the child. Adapting and contextualising existing
block an individual and report inappropriate
evidence-based programmes should be prioritised
content or requests. Furthermore, establishing
and existing evidence-based materials considered as
children’s knowledge on the risks inherent to
a starting point.
online interaction and the exchange of personal
For awareness and education programmes in information, images and videos.
schools, the Ministry of Education and Kenya 1.2. When children do not know about sex, it
Institute of Curriculum Development have a leading enables offenders to take advantage. We must
role to play. For reaching wider communities across ensure that knowledge reaches all children, and
the country the Communications Authority of Kenya include information about sex, consent, personal
and the Department of Children’s Services, with its boundaries, what adults or others around children
presence at the local level, are important institutions. can or cannot do to them, risks and responsibilities
The Ministry of Information Communication and when taking, sending and receiving sexual images,
Technology, the Anti Human Trafficking and Child and how to say no to others. This will help children
Protection Unit of the Department of Criminal to identify risky or inappropriate interactions both
Investigations, the Kenya Film Classification Board online and in person.
and National Government Administration Officers
(NGAO) can all support these efforts with their own 1.3. Adjust education and awareness raising
expertise and in their respective sectors. approaches to reach children. Pamphlets may
not reach the intended audience. It is essential to
Liaising with churches, child rights clubs and peer employ child-friendly language and engage through
mechanisms can also be important to create awareness social media and messaging platforms that children
about OCSEA (including reporting mechanisms – see are using. Campaigns should be designed and run
Insight 2). Information on OCSEA should be included in by people who understand the approaches and
parenting education programmes. Special care should messages; they must receive experiential training
be taken to ensure that information is communicated to and support to be able to design and share this type
children with disabilities.
156. See: eSafety Commissioner’s programme: ‘Start the Chat’.

157. See: eSafety Commissioner’s programme: ‘Be Connected’
of material. Better still are campaigns and messaging the risk and benefits of being online and lead to a
that involve young people in their design and more open dialogue between children and adults
delivery. when children face dangers or harm online.
Existing tools like the Ministry of Education’s online 1.5. Inform children about their right to be
safety and security manuals may serve as one protected from all forms of physical, sexual, and
avenue for delivering the messages identified in the emotional abuse and exploitation, and on how
Disrupting Harm research. However, ensuring that to stay safe by setting boundaries, recognising
these issues are explored with the necessary depth appropriate and inappropriate behaviour from adults
and sensitivities also requires training and support for and those around them and how to say
those required to deliver such complex material. no to inappropriate behaviour.
Caregivers, teachers, medical staff and social 1.6. Caregivers and duty bearers should learn about
support services what children are doing online and offline, and be
1.4. Improve understanding of digital platforms vigilant about the people that their children or the
and technologies. Around half of the caregivers of children in their community interact with. Consider
internet-using children in Kenya have never used the whether these interactions seem appropriate for
internet themselves. Being involved and supportive children. Only some threats come from strangers on
of a child’s internet use will help them understand the internet.
Disrupting Harm alignment with the evidence from Kenya shows that even though
Model National Response many of the capabilities in the Model National
Many countries, companies and organisations Response exist, they are not functioning optimally.
have joined the WePROTECT Global Alliance Our recommendations primarily address
to prevent and respond to online child sexual legislation,158 dedicated law enforcement,159
exploitation and abuse. judiciary and prosecutors,160 and education
As a member of the Global Alliance, Kenya can use programmes.161 All recommendations are
the Model National Response to Preventing and practical, evidence-based and actionable.
Tackling Child Sexual Exploitation and Abuse to Disrupting Harm has also indicated to whom its
help organise its response to OCSEA. The Model is various recommendations are addressed – i.e.,
a valuable tool for governments to organise and government duty-bearers, law enforcement
improve the level of their response. authorities, justice professionals, the internet and
technology industry, or caregivers, the community
Most of the recommendations in this report align and teachers.
with the 21 ‘capabilities’ articulated in the Model
National Response, but Disrupting Harm identifies The recommendations are organised under five
priority areas for interventions based specifically key insights from the Disrupting Harm evidence,
on the data about the situation in Kenya. The and signposted for different stakeholder groups.
158. Model National Response #3.

INSIGHT 2 A further consideration from the data
Children abused by an offender of the same
Many children in Kenya did not sex may have difficulty disclosing instances
tell anyone the last time they were of exploitation or abuse or seeking help due
to the stigma associated with being viewed
subjected to OCSEA. Children tend to
as homosexual which involves strong societal
disclose to people they know rather taboos, and is criminalised. Children may fear
than reporting to a helpline or the legal consequences if they report. Although our
police. survey results show that boys and girls are both
subjected to OCSEA, no male victims could be
identified for interview during the research for
Government
Disrupting Harm in Kenya.
2.1. Provide public financial support to Childline
Kenya in order to ensure its sustainability and
improve its ability both to receive reports and to
provide psychosocial support to children. Allocations
should be made for this purpose in the budget of Law enforcement
the Ministry of Labour and Social Protection. The 2.4. Establish a clear reporting process for cases
Ministry of Health, the Communication Authority of OCSEA and facilitate widespread training for
of Kenya and internet service providers (Safaricom, all police and other duty-bearers to ensure full
Airtel) could also contribute. However, voluntary implementation, so that children and families
contributions should only be complementary. In are comfortable about reporting instances of
return, Childline Kenya could be requested to assess abuse. Ensure that child-friendly procedures are
their efficiency and hold extensive consultations with implemented whenever children are involved as
children on how to best provide support for OCSEA. victims so as to make the justice process a more
positive experience. Ensure that the Justice for
2.2. Increase awareness raising efforts about
Children strategy being developed by the National
hotlines and helplines as a reporting and help-
Council on the Administration of Justice includes
seeking mechanism for OCSEA. An important
OCSEA-related issues and child-friendly reporting
prerequisite is that helplines should be adequately
procedures.
resourced and providing good quality care and
support. Even if children are made aware of 2.5. Create formal mechanisms for the sharing of
helplines, if initial responses to disclosure and help- evidence from OCSEA cases between civil society
seeking are poor, the child – and others observing organisations and the police. This will ensure
the case – will be much less likely to seek help again. that frontline workers are not forced to develop
workarounds which may be problematic or illegal.
2.3. Diversify mechanisms for children to disclose
concerns, seek help and formally make reports
(including simple child-friendly, online methods),
bearing in mind that most children first prefer to
seek help from friends, then within their own family
or community.
Caregivers, teachers, medical staff and social INSIGHT 3

support services
2.6. Foster safe and ongoing communication
between children and trusted adults about their lives
Among children who were subjected to
online. Normalising communication about online OCSEA through social media, Facebook
activities will increase the likelihood that children will and WhatsApp were the most common
disclose any concerns, risks and harmful experiences platforms where this occurred.
they may face.
2.7. Responses to disclosures of OCSEA should Law enforcement

always convey that it is never the child’s fault, 3.1. Improve law enforcement officers’ abilities
whatever choices they have made. It is always the to flag/refer cases of OCSEA to global technology
fault of the adult abusing or exploiting the child. Our companies.
research shows that children subjected to OCSEA
often blame themselves and feel that they had let 3.2. Improve officers’ ability to report content hosted
their caregivers and others down, or were judged by outside of the country – e.g., on a website. This is most
the police. Responses should be without judgement often achieved by an effective hotline having access
or punishment. For example, see guidelines on first to hotlines in other countries (e.g., via the INHOPE
line response to child maltreatment. network), which can then serve a take-down notice on
their domestic internet service providers.
2.8. Try not to restrict children’s internet access as
a response to potential harm. Restricting access to Government
technology is seen as a punishment. It only protects 3.3. Impose legal duties on internet service providers to
children temporarily and does not teach them retain data for a set minimum period and to filter and/or
how to navigate similar situations in the future. block and/or take down CSAM as well as to comply with
This response also tends to discourage children law enforcement requests for information in a prompt
from confiding in adults about the problems they manner. This will assist investigations into crimes as well
experience. as aid in controlling the wide distribution of CSAM.
2.9. Invest in improving the capacity of social 3.4. Impose legal duties on social media platforms
service workforce. Improve capacity of frontline staff to ensure that there is a strategic and well-funded
in contact with children to better identify children effort to minimise children’s experiences of OCSEA
at risk or that have experienced OCSEA. This should on their platforms. Hold social media platforms
include teachers/pastoral care staff in schools as well legally responsible for facilitating cases of harm
as health workers, additional to all those providing against children.
psychosocial support (see insight 4).
Industry
2.10. Help children understand the full extent of the 3.5. Make formal reporting mechanisms within
risks of sharing sexual content and how to engage platforms clear and accessible to children and
in harm minimisation to limit possible negative detail in child-friendly terms what the process looks
repercussions. Most children who shared sexual like after children submit a report. Platforms and
content initially did so because they were in love or service providers must respond rapidly to reports
trusted the other person, but this behaviour can lead made by children and demonstrate transparency and
to serious harm, such as non-consensual sharing of accountability. Platforms should also work proactively
the content with others and sexual extortion. to prevent sexual content from appearing on children’s
feeds and where relevant adhere to government
regulations on how to do so.
3.6. Internet service providers should comply with

regulations to filter and remove CSAM. Enforcing this
action is vital in keeping children safe online.
INSIGHT 4 4.6. Operationalise the Victim Protection Trust
Fund to provide victim support services (including
counselling and medical care) where there are no pro
The law enforcement, justice and social bono services available, and to meet victims’ logistical
support systems have inadequate expenses during the criminal justice process.
awareness, capacity and resources to 4.7. Expand access to legal aid to ensure more
respond to cases of OCSEA. child victims go through the justice system with a
lawyer. The Children Act entitles every child to legal
Government support at the expense of the State. To achieve this,
4.1. Urgently invest in the training of police officers, however, both funding and the awareness of victims
prosecutors, judges/magistrates, lawyers, courtroom of their right to legal aid need to be increased. The
staff, child protection officers and frontline workers Department of Children Services may coordinate.
on what OCSEA is and how to address it within their Law enforcement
respective professions. Address child protection
4.8. Improve data collection and monitoring of
issues including OCSEA in basic training and provide
OCSEA cases both in the AHTCPU and all police
specialist training more widely. Provide both initial
stations in Kenya. It is advised to identify and record
and refresher training.
OCSEA indicators (i.e. was there a technology element
4.2. Provide adequate and sustainable funding involved) in all case records related to child sexual
for all agencies involved in tackling OCSEA such as exploitation and abuse at both the AHTCPU and
the Department of Children’s Services, Anti-Human police stations in Kenya with breakdowns of victim’s
Trafficking and Child Protection Unit, and Childline and offender’s characteristics.
Kenya.
4.9. Train all police officers and prosecutors,
4.3. Support the National Police Service in especially at the county and sub-county levels, about
establishing more specialised child protection units the linkages between online and in-person forms
with trained female and male personnel capable of of child sexual exploitation and abuse. Inform them
delivering child-friendly support, and the physical about the provisions of law that can be used to bring
spaces and equipment needed to do so. Increase charges in cases of abuse in the online environment.
the expertise, resources and staff of the AHTCPU to
4.10. Guidelines for police officers should be adapted
ensure a presence in more counties/within the 11
to incorporate best practice on interviewing
regions. Strengthen the links between the AHTCPU
children during the criminal justice process. This will
and the police stations at the county level for
prevent children from being interviewed repeatedly,
guidance. Make sure that officers trained in handling
which can feel like a form of secondary victimisation.
OCSEA cases are not transferred to other units
As in the AHTCPU, investigators could record the
without a suitable replacement.
interviews and share a copy of the interview with
4.4. Provide psychological support to all those the prosecutor and the court instead of arranging
working with victims of OCSEA, including probation multiple interviews.
officers, Department of Children’s Services workers,
4.11. Ensure that police officers/prosecutors/courts
prosecutors, magistrates, lawyers, social workers,
have a standard information package to provide to
mental health professionals and personnel of the
all victims and their caregivers related to child sexual
AHTCPU.
exploitation and abuse (including OCSEA) to ensure
4.5. Ensure that the arrangements for child- that all the relevant procedures and rights, including
friendly justice envisaged in the Children Act are their right to compensation, are clearly explained.
implemented consistently in all cases of child sexual The National Council of Children’s Services could be
exploitation and abuse crimes, including those with mandated to develop such a package. This will enable
online elements. This will require financial resources, child victims to make informed decisions as well as
operating procedures and training. familiarise them with the upcoming procedures.
4.12. Maintain the connection to INTERPOL’s Child 4.19. Request a victim impact statement for OCSEA
Sexual Exploitation (ICSE) database. Improve the cases. This will help create awareness of the impact
availability and speed of the internet connections of OCSEA and allow the victim to feel truly seen and
at the Nairobi Anti Human Trafficking and Child heard in the court process.
Protection Unit (AHTCPU) and connect the Mombasa
AHTCPU to the internet and the ICSE database. Social support services
Ensure the units have the necessary Triage tools to 4.20. Train all staff on the frontline of social support
deal with CSAM. Mobilise resources with the Ministry services (not just specialist services) to recognise the
of Interior in conjunction with the Treasury to sustain unique risks and harms of OCSEA, and provide them
the AHTCPU project. with evidence-based best practices for responding.
When children are brave enough to seek help, those
4.13. Provide an effective mechanism and adequate they seek help from must be equipped to provide it.
resources to ensure that international OCSEA
referrals, including NCMEC CyberTips, are subject to 4.21. Social support services need to find modern
an appropriate level of investigation, with a view to and innovative ways of being accessible to
minimising ongoing harm to children. young people. Helplines are one way of achieving
widespread access to a child population. These
4.14. Law enforcement to strengthen collaboration need substantial investment and resourcing: their
with social services. Law enforcement should seek mere existence is not sufficient. Other social support
to collaborate with social services whenever possible, services need online means of access, and support
to ensure a victim-centered justice process. from young, trained staff who understand the way
children engage in their online lives.
Justice professionals
4.15. Train all justice actors, including prosecutors Industry
and judges, on how to handle OCSEA cases and 4.22. Prioritise responding to data requests from
deliver child-friendly justice as dictated by the the courts in cases involving children to help reduce
Children Act. the duration of trials.
4.16. Limit the duration of criminal court cases
that include child victims. OCSEA cases must be
processed and adjudicated without undue delays A further consideration from the data
to secure digital evidence and protect the child’s During the Disrupting Harm research activities,
wellbeing. Courts could grant priority to cases respondents expressed concern that requests
involving children when scheduling hearings, or the for informal payments and corrupt influences
Children Act could be amended to limit the duration on judicial processes constitute serious barriers
of cases. to formal reporting and obstruct access to
justice in some instances.
4.17. Develop and implement programmes
preparing the child victim to engage with the court
system and legal actors.
4.18. Ensure that child victims do not have to face

the offender – for example, by employing video-
link technology so that evidence may be given
from another room. The court methods used in
the Barnahus model162 may also be explored for
adoption. If these options are unavailable, witness
protection boxes can be used (although boxing the
offender rather than the child is preferable).
162. See: Child-friendly centres for abuse victims: Barnahus.
INSIGHT 5 5.5. Ensure data retention and preservation policies
are in place to support law enforcement authorities
in criminal investigations.
Important OCSEA-related legislation,
5.6. Adopt and enforce the upcoming Children
policies and standards are not yet
Bill 2021 to properly define and criminalise OCSEA
enacted in Kenya. in law.
Government 5.7. Amend legislation in such a way as to extend

the crime of online grooming for sexual purposes
5.1. Integrate OCSEA in other policies addressing
to situations where the sexual abuse is not the result
violence against children.
of a meeting in person but is committed online (e.g.,
5.2. Prioritise the ratification of the Optional when children are manipulated to produce CSAM
Protocol to the Convention on the Rights of the and share it with the offender).163
Child on the sale of children, child prostitution
5.8. Amend legislation to explicitly criminalise the
and child pornography and amend legislation to
live-streaming of child sexual abuse and sexual
bring it fully into line with the standards set by the
extortion committed in the online environment. The
Protocol. This Protocol is relevant to combating CSAM
Computer Misuse and Cybercrimes Act is a milestone
and other crimes related to the sexual exploitation of
in Kenya’s fight against OCSEA, as it provides the
children. Kenya signed the Protocol in 2000 but is yet
procedural rules needed to assist law enforcement
to indicate its consent to be bound by its provisions
officers in the investigation of OCSEA cases. However,
by ratifying it. The Office of the Attorney General, the
it does not explicitly criminalise knowingly obtaining
Kenya Law Reform Commission and the National
access to CSAM or any forms of OCSEA other than
Council for the Administration of Justice need to
conduct related to CSAM.
be engaged with respect to the amendment of
legislation, including all legislation mentioned below. 5.9. Take the necessary steps to support the
effective implementation of the National Plan of
5.3. Accede to the Convention on Cyber Security
Action Against Sexual Exploitation of Children in
and Personal Data Protection adopted by the
Kenya, 2018-2022 and the National Information,
African Union in 2014. With respect to OCSEA, the
Communications and Technology Policy (2019).
Convention specifically includes CSAM.
The National Council of Children’s Services and
5.4. Consider amending legislation to conform Department of Children’s Services can take the lead
to other international conventions which offer on this. These steps should include the dissemination
good guidance for addressing OCSEA, such as the of these policies to relevant implementing agencies
Council of Europe’s Convention on the Protection and the allocation of the budgets needed for
of Children Against Sexual Exploitation and Sexual implementation and regular monitoring of progress.
Abuse (Lanzarote Convention) and Convention
5.10. Ensure that the programmes and solutions
on Cybercrime (Budapest Convention). These
foreseen by the National Plan of Action on online
conventions provide useful measures of national
child sexual exploitation and abuse and the
legal frameworks related to OCSEA and are open for
National Strategy on Child Online Protection
accession by states which are not members of the
complement each other.
Council of Europe.
163. The Lanzarote Committee recommended States Parties to the Convention to adopt this broader understanding of online grooming in its 2015
opinion on Article 23 of the Convention: Council of Europe’s Lanzarote Committee. (2015). Opinion on Article 23 of the Lanzarote Convention and its
explanatory note. Para 20.
ACKNOWLEDGEMENTS
ECPAT, INTERPOL, and UNICEF Office of Research Supporting data collection: Child Helpline
– Innocenti have greatly appreciated the unique International, Directorate of Criminal Investigation
opportunity to work shoulder-to-shoulder to assess – Anti-Human Trafficking and Child Protection Unit
OCSEA in Kenya. This comprehensive report is the (AHTCPU), the Kenya Alliance for Advancement of
result of a two-year collaborative effort to design Children (KAACR), the International Association
research, gather data and produce extraordinary for Internet Hotlines (INHOPE), the Internet Watch
evidence. These efforts would not have been Foundation (IWF), INTERPOL National Central Bureau
successful without the engagement of so many (NCB) Nairobi, Ipsos Kenya, Ipsos MORI, Missing
individuals and partners in Kenya. The project Child Kenya, Awareness Against Human Trafficking
partners would like to express their appreciation to (HAART), UNICEF Eastern and Southern Africa
everyone who engaged with Disrupting Harm by: Regional Office, UNICEF Kenya.
Contextualising the findings: The African Sharing expertise and experiences through
Union, Child Fund Kenya, Childline Kenya, the interviews and survey responses: African Network
Communication Authority (KE-CIRT and Consumer for Prevention and Protection of Child Abuse and
Affairs Division), CRADLE, the Department of Neglect, Childline Kenya, the Child Rights Network,
Children’s Services of the Ministry of Labour and Child's Life, the Coalition on Violence Against
Social Protection, the Directorate of Criminal Women, community volunteer officers, CRADLE,
Investigation – Anti-Human Trafficking and Department of Children’s Services, Directorate of
Child Protection Unit (AHTCPU), Eveminet Criminal Investigations, Eveminet Communications
Communications Solutions Limited, HAART Solutions Limited, Facebook, Hope in Life, the
Kenya, Hope in Life, the Kenya Alliance for Kenyan Police – Child Protection Units, Kesho Kenya,
Advancement of Children (KAACR), the Kenya Ladder of Hope Kenya, Love Justice International,
Films Classification Board, the Kenya Institute of members of the Judiciary, International Justice
Curriculum Development under the Ministry of Mission, Missing Child Kenya, Mtoto News, Solidarity
Education, the Kenya Law Reform Commission, with Girls in Distress, Stop the Traffik Kenya, teachers
Ladder of Hope Kenya, Love Justice International, of the Ministry of Education, Terre Des Hommes
the Ministry of ICT, Information and Youth Affairs, Netherlands, Trace Kenya, UNICEF Kenya, Watoto
Missing Child Kenya, Mtoto News, the National Watch Network,
Council for the Administration of Justice – Special
Taskforce on Children Matters, the National Council Our biggest thanks go to the children who
for Children's Services, the Office of the Director of contributed – especially the young people who
Public Prosecutions, the Judiciary, Plan International had experienced OCSEA and courageously spoke
Kenya, Save the Children Kenya, Solidarity with Girls of it with the research teams. The experiences of
in Distress, Terre des Hommes, Trace Kenya, UNICEF children are key to understanding and guiding our
Eastern and Southern Africa Regional Office, UNICEF way forward.
Kenya, Watoto Watch Network. The partners also acknowledge the guidance
of the Panel of Advisors and the extraordinary
financial investment in this project from the Global
Partnership to End Violence against Children
through its Safe Online initiative. We are grateful
to the Safe Online team for its conceptualisation of
Disrupting Harm, its technical contributions and its
unwavering support.

DH Kenya Report

Uploaded by

Copyright:

Available Formats

DH Kenya Report

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

DH Kenya Report

Uploaded by

Copyright:

Available Formats

DISRUPTING

1. Children Online in Kenya 22

1.1 Internet access and barriers 23

2. Online Child Sexual Exploitation and Abuse in Kenya 36

2.1 Law enforcement data 38

3. Responding to Online Child Sexual Exploitation and Abuse in Kenya 70

3.1 Formal reporting mechanisms 71

4. How to Disrupt Harm in Kenya 92

Our online lives are advancing constantly. The internet and

With progress being made towards universal internet connectivity, it is ever-more

Disrupting harm to children is everyone’s responsibility.

Perceptions of online risks Although these encounters tended to be positive for

While most caregivers say they would tell the police

National Consultation were asked to comment on the Disrupting Harm

Figure 1: Disrupting Harm methods in Kenya

survey data professionals n=9

Desk review of relevant documents

Child sexual abuse refers to various sexual activities perpetrated on children

Sexual exploitation Child sexual

POPULATION TOTAL 2019 FEMALE POPULATION 2019 MALE POPULATION 2019

52,574,000 26,452,000 26,122,000

(2018: 51,393,000) (2018: 25,859,000) (2018: 25,534,000)

Census data: Census data: Census data:

47,564,000 24,015,000 23,548,000

2030 prospect: 33.4%29

Source: Disrupting Harm data

MOBILE (SIM) PENETRATION

n = 1,879 households. Home

GDP PER CAPITA 2019 (US$) 20% 27% 99%

ONE OF THE FASTEST GROWING ECONOMIES Mail

THE INEQUALITY GAP (HIGHEST

2014 2015 2017

HUMAN DEVELOPMENT INDEX The Human Development 41

Index (HDI) is a summary OFFICIAL LANGUAGES

Source: Disrupting Harm data

FREQUENCY OF INTERNET USE AMONG 12-17 YEAR OLDS

40% Less than once a month

n = 1,014 internet-using children.

FREQUENCY OF INTERNET USE AMONG

CAREGIVERS OF INTERNET-USING CHILDREN

ICT DEVELOPMENT GLOBAL

Source: Disrupting Harm data

n = 1,014 internet-using children.

MARKET SHARES IN MOBILE DATA

Figure 3: Frequency of children’s internet use

Total 27% 13% 33% 27%

12–13 32% 15% 33% 20%

14–15 25% 13% 39% 23%

16–17 26% 11% 28% 35%

Boy 22% 13% 35% 30%

Girl 32% 13% 30% 24%

0% 20% 40% 60% 80% 100%

Figure 4: Frequency of caregivers’ internet use

Total 48% 12% 3% 9% 28%

< 29 22% 13% 4% 14% 47%

30-39 48% 13% 3% 8% 28%

40-49 52% 9% 2% 9% 27%

50+ 73% 15% 3% 3% 6%

Men 44% 13% 2% 7% 33%

Women 50% 11% 3% 10% 25%