I was looking around to see if wedge attacks on EMV cards are still viable or a solution has been implemented, and if so, have attackers found another way around the solution?

The latest literature or post I can find goes back to 2010-2012.


1 Answer 1


Wedge attack was applicable on DDA cards (Dynamic Data Authentication) which were being superseded by CDA cards (Combined Data Authentication) at the time when vulnerability was disclosed.

The researchers behind the vulnerability proposed a solution to fix DDA cards in his write-up: Defending against wedge attacks in Chip & PIN but there's no track record of which banks globally enforced this solution or even cared to address it.

CDA partially resolved the problem with DDA but it was itself open up to a vulnerability called No PIN attack. The wedge was used even in this attack but the method of exploitation was different from that in DDA as explained in their research paper, Chip and PIN is broken.

Their proposed solution as mentioned in the paper fixes CDA.

One possible work-around is for the terminal to parse the IAD, which does include the result of PIN verification. The card can change its CDOL to request that the CVMR be included in the payload to the Generate AC command.

Unless the CVMR is included in the CDOL it may not be integrity protected, so a second man-in-the-middle between terminal and acquirer (perhaps installed with co-operation of a corrupt merchant staff member) could tamper with it too.

Barclays Bank UK PLC (52:20) and HSBC Bank (34:28) are known to block No PIN attack. They were able to block it without rolling out any fixes by comparing Card Verification Results (CVR) stored in Issuer Application Data (IAD) with Cardholder Verification Method Results.

EMV v4.4 Book 3 Application Specification, CVR Byte 2 Table CCD 10: Card Verification Results for Format Code 'A'

EMV v4.4 Book 4 Other Interfaces, A4 Table 33: CVM Results

EMV Book 4, 12.1.1 - Authorization Request - Table 9 contains the new data elements specifically created for an ICC transaction. It is added in the book in November 2011, a year after the vulnerability is disclosed. Now the terminal can optionally send CVMR (Cardholder Verification Method Results, specifies which cardholder verification method the terminal believes was used) with the transaction approval request but unlike IAD (Issuer Application Data), it it is not integrity protected. The issuer then validates CVMR against IAD to ensure both the card and terminal have the same perspective on how the transaction is happened.

A year after the vulnerability is disclosed publicly, a criminal ring was caught performing this attack which the Cambridge researchers believe that the French criminals worked out the attack independently.

The ENS and CEA forensic researchers note that the vulnerabilities used by the French fraud they analyzed have since been fixed—at least in Europe—though they declined to fully detail the new security measures.

The French researchers who investigated the case write in their paper that EMVCo has since created new countermeasures to the vulnerabilities the fraudsters exploited and implemented them both in card readers and in banking networks. They also note that other protections have been added to the system at the network level, which they decline to detail for fear of tipping off criminals.

I doubt all the banks globally have implemented this solution as another decade old vulnerability is again discovered to be lurking around in some banks: Researchers create magstripe versions from EMV and contactless cards: Banking industry loophole reported more than a decade ago still remains open and ripe for exploitation today.

The only reason they are not seeing frauds is because most of them enforce online PIN verification. But as the specification was updated in 2011 and most banks outsource the production of EMV cards, they might be coming with a fix particularly the new ones in countries where EMV is recently mandated.

  • That’s a great answer. I don’t really have much to add other than to note that some banks said they had ways to detect the no-PIN attack that didn’t need any new data fields. I never found out any details and suspect it’s not perfect but was considered good enough to reduce the risk while the proper fix gets (slowly) rolled out. Oct 6, 2020 at 8:35

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .