Does The Online Card Payment Landscape Unwittingly Facilitate Fraud?
Does The Online Card Payment Landscape Unwittingly Facilitate Fraud?
Does The Online Card Payment Landscape Unwittingly Facilitate Fraud?
Copyright:
2017 IEEE. Personal use of this material is permitted.
Permission from IEEE must be obtained for all other uses, in any current or future media, including
reprinting/republishing this material for advertising or promotional purposes, creating new collective
works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work
in other works.
Date deposited:
01/12/2016
attack;
I. INTRODUCTION
Sales and fraud statistics from regions other than the UK are less reliable but
indicate the same pattern.
Fig. 2. Screenshot of the website bot, farming CVV2 from multiple sites.
these sites to guess the expiry date. 291 sites use three fields,
which one can use for guessing the CVV2, and 25 sites use four
fields, which allows one to guess the postcode of the address.
Finally, of the 389 sites, 47 merchants (i.e. 12%) had
implemented 3D Secure payments (these sites are impervious to
the distributed guessing attack, see Section VI.B).
Sites
with 3
fields
(guess
CVV2)
Sites
with 4
fields
(guess
address
postcode)
Sites
with 3D
Secure
(safe
from
attack)
Total
0 to 5
23
27
6 to 10
20
238
18
276
11 to 50
28
33
Unlimited
3D
Secure
47
47
Total
26
291
25
47
389
V. RESPONSIBLE DISCLOSURE
Two weeks after we completed the distributed guessing
attack experiments, we initiated an ethical/responsible
disclosure exercise, notifying Visa and a selection of affected
sites. Based on the number of fields that a website checks, we
categorised them into three groups: expiry date, CVV2 and
postcode. Since the total number of vulnerable websites was
very high, we selected the 12 biggest players from each
category (in terms of the highest number of users), taking the
total number of notified websites to 36.
Once a suitable contact person or team for each website was
found, we presented them with the disclosure information that
featured the experiments we performed and the type of
vulnerabilities on their site. We used our official work/university
email address and this served as a means for these merchants to
trace us back, so that they can verify our authenticity. This
would also allow them to request more detailed and technical
information about our experiments should they wish to find out
more.
We recorded the responses received from these websites
over the duration of four weeks after we disclosed the
vulnerabilities to them. Altogether, we received 20 human
responses from 10 websites and 18 websites came back to us
with machine generated response mostly confirming the receipt
of our notification. All of the human responses requested more
technical details while some asked us to suggest solutions. Out
of the 36 websites we contacted, eight never responded. When a
web merchant requested more information, we offered them an
initial draft of this article, which explained the experiments and
the attack to help them understand the actual problem. We
followed the disclosure policy requested by the websites and
anonymised the affected sites in our article.
TABLE II.
Web
site
Informa
tion
Leak
Adding
Addr.
field
Adding
Delay
filter
Adding
velocity
filter
(PAN
based)
Adding
velocity
filter (IP
based)
Adding
CAPTC
HA
Exp. date
Exp. date
Exp. date
Exp. date
CVV2
CVV2
CVV2
CVV2
habits [6]. The issuing bank then has the option to block the
payment, or ask the customer for confirmation, or accept the
payment taking a calculated risk that a transaction may be found
to be fraudulent later. A complicated set of considerations comes
to the fore in the banks decisions, from ease of use to financial
risks. However, one would expect that if they so desire, banks
could have considerable influence on the payment gateways and
card payment networks in protecting against the distributed
guessing attack.
[3]
[4]
[5]
[6]
VII. CONCLUSION
In this paper, we studied 400 of the most popular ecommerce websites and surveyed their web payment interface,
identifying that different websites present different sets of fields
to identify the cardholder. It turns out that this disparity between
different websites inadvertently creates conditions for a scalable
distributed guessing attack. By conducting a guessing attack one
field at the time using a set of appropriate websites at each
stage the attack becomes practical. With the obtained data, the
attacker can make purchases or transfer funds, as we have
demonstrated.
We showed that the attack works if the card payment
network is not able to relate card activities from different
websites. Fundamentally, much of the problem with card
payment stems from the fact that the identity of the payer needs
to be established in the card-not-present mode. This is
inherently problematic since it is at odds with the original use of
cards (where the card and cardholder are present at the moment
of purchase). It also implies that, for instance, Chip-and-PIN is
not available to establish the identity of the payer. This is
exacerbated by the fact that the Internet facilitates distribution
of guesses for data fields over many merchant sites.
To prevent the attack, either standardisation or centralisation
can be pursued (some card payment networks already provide
this). Standardisation would imply that all merchants need to
offer the same payment interface, that is, the same number of
fields. Then the attack does not scale anymore. Centralisation
can be achieved by payment gateways or card payment networks
possessing a full view over all payment attempts associated with
its network. Neither standardisation nor centralisation naturally
fit the flexibility and freedom of choice one associates with the
Internet or successful commercial activity, but they will provide
the required protection. It is up to the various stakeholders to
determine the case for and timing of such solutions.
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
American
Express
SafeKey,
Product
Capability Guide,
https://network.americanexpress.com/en/globalnetwork/Images/SafeKey
ProductCapabilityGuide_2014.pdf, Accessed: 04 Mar 2016.
Visa,
3D
Secure
System
Overview,
http://www.visanet.com.pe/verified/demovisanetweb/resources/3DS_70015-01_System_Overview_external_v1.0.3.pdf,
2001, Accessed: 12 May 2016.
MasterCard, MasterCard Secure Code, Merchant Implementation
Guide,
https://www.mastercard.us/content/dam/mccom/enus/documents/SMI_Manual.pdf, 2014, Accessed: 13 May 2016.
Visa, Card Acceptance Guidelines for Visa Merchants,
http://usa.visa.com/download/merchants/card-acceptance-guidelines-forvisa-merchants.pdf, 2014, Accessed: 13 May 2016
Payment Card Industry, PCI DSS Applicability in an EMV
Environment,
https://www.pcisecuritystandards.org/documents/pci_dss_emv.pdf,
2010, Accessed: 13 May 2016.
PayPal,
Gateway
Developer
Guide
and
Reference,
https://www.paypalobjects.com/webstatic/en_US/developer/docs/pdf/pa
yflowgateway_guide.pdf, 2014, Accessed: 11 Mar 2016.
MasterCard,
Transaction
Processing
Rules,
http://www.mastercard.com/us/merchant/pdf/TPREntire_Manual_public.pdf, 2014, Accessed: 13 May 2016.
BinDb, Bank Identification Numbers Database - Credit Card Bin
Lookup, https://www.bindb.com/structure.html, Accessed: 13 May
2016.
ExactBins, Exact BIN Database, https://www.exactbins.com/features,
Accessed: 13 May 2016.
Alexa,
Alexa
Top
Shopping
Sites,
http://www.alexa.com/topsites/category/Top/Shopping, Accessed: 13
May 2016.
SeleniumHQ,
"Selenium
framework
documentation.
http://docs.seleniumhq.org/docs/, Accessed: 17 May 2016.
M. Emms, B. Arief, L. Freitas, J. Hannon, and A. van Moorsel,
Harvesting High-Value Foreign Currency Transactions from EMV
Contactless Credit Cards Without the PIN, In Proceedings of the 2014
ACM SIGSAC Conference on Computer and Communications Security
(CCS14), ACM, 2014, pp. 716-726.
Symantec, Validating a Credit Card Number using Luhns Algorithm.
https://support.symantec.com/en_US/article.TECH221769.html
[Accessed: 09 May 2016]
A. El Ahmad, J. Yan and W. Ng, "CAPTCHA Design: Color, Usability,
and Security", IEEE Internet Computing, vol. 16, no. 2, pp. 44-51, 2012.
Adyen, "Analysis Reveals Worldwide Impact of 3D Secure on
Transaction Conversion Rates, https://www.adyen.com/home/aboutadyen/press-releases/2014/3d-secure-worldwide-impact-conversion,
2014, Accessed: 13 May 2016.
MasterCard,
The
MasterCard
Network
Advantage,
http://newsroom.mastercard.com/wpcontent/uploads/2011/09/MasterCard-Network-Advantage.pdf,
Accessed: 12 May 2016
ACKNOWLEDGEMENTS
[2]