Slashdot Mirror
Chip and Pin "Weakness" Exposed By Cambridge Researchers
another random user writes "A vulnerability in the widely used chip and pin payment system has been exposed by Cambridge University researchers. Cards were found to be open to a form of cloning, despite past assurances from banks that chip and pin could not be compromised. In a statement given to the BBC, a spokeswoman for the UK's Financial Fraud Action group said: 'We've never claimed that chip and pin is 100% secure and the industry has successfully adopted a multi-layered approach to detecting any newly-identified types of fraud.'"
Lots of these systems use proprietary protocols and have pushed out 3rd party verification by researchers. the random number being generated by time? Any serious security auditor would have caught that if the banks allowed them in, one of the golden rules of cryptography is to have a proper random number generator. The contact-less systems in the US came under similar fire this past year, after years of assurances by card issuers that it couldn't happen. http://www.forbes.com/sites/andygreenberg/2012/01/30/hackers-demo-shows-how-easily-credit-cards-can-be-read-through-clothes-and-wallets/
The man who cannot imagine a horse galloping on a tomato is an idiot - Andre Breton
in DEF CON 19 last year?
All the locks in the world won't keep crooks out of your house if you don't use the locks. Your house may LOOK invulnerable, but one day sonbody's gonna try the door, find it open, and steal you blind.
The same principle applies here - using obvious and predictable 'random' code generation, and relying on people not knowing that's what you're doing, only works for so long.
And arrogant people, (and companies, and banks), who crow about how secure their systems are, are just asking for it. Serves the fuckers right; but it's too bad that credit card holders are paying the price for their creditors' arrogance.
'The Economy' is a giant Ponzi scheme whose most pitiable suckers are the youngest among us and the yet-unborn.
In the US, a simple magnetic stripe is used to encode the data, which can be duplicated with little effort. Even if your credit card is swiped at a brick and mortar retailer, this well-known vulnerability gives consumers some credibility against the credit card issuer when they claim to have not made the purchase. The scary part of this chip and pin vulnerability is that banks have a history of blaming the consumer and not issuing refunds since chip and pin was presumed to be secure. From the article, "Others [banks] reported already being suspicious of the strength of unpredictable numbers... If those assertions are true, it is further evidence that banks systematically suppress information about known vulnerabilities, with the result that fraud victims continue to be denied refunds."
Canadian banks just snuck in an update to the banking agreements--customer is now 100%responsible for losses with chip and pin cards, no doubt due to the ironclad security.
The problem with the claim Chip & Pin is more secure, is that the card processors (Visa, Mastercard) used it as a justification to shift liability from the Bank over to the Merchant.
With swiped transactions, when a customer disputes the transaction, the Merchant isn't automatically liable for the transation -- they only need to prove the customer actually made the purchase (e.g. producing the signed receipt). With Chip & Pin, the merchant is automatically assumed to be liable, according to the merchant agreement. There's very little a merchant can do to dispute the chargeback.
Yeah, they pass it along to sellers like me. Almost all fraud gets taken straight out of the pockets of the business owner but hey, we've got money, right? Total bullshit. Well guess what I'm refusing to accept ever under any circumstances.
I like how they highlight "weakness" in the headline, giving it the appearance of being of poor credibility. Can I try?
BBC is a "news" provider.
doesn't even have a chip. It's from a major American bank.
This appears to be something new, however
File under 'M' for 'Manic ranting'
The crooks have been cloning chip and pin cards for a couple of years now, why is it that it only becomes news when Cambridge Researchers catch up and realise the same thing ?
You can buy the gear you need to do it for about £200 from online retailers no questions asked.
I realised how frail and easily crackable the entire system was when I was asked to research an EPOS system using an online payment gateway setup and discovered that at certains points in the system all the information exists in plain text, the onus is on the developer to ensure the data remains encrypted.
Folks, I read the paper by Omar and Co in a fair amount of detail. Here is the gist. Some ATMs do not have a true RNG (Random Number Generator), something like FIPS 140.2 compliant. With such defective systems in a particular country, at a particular time and for a particular amount and a system which can do a transaction at mS granularity accuracy an attack is possible. And the card has to be in the system (which is recording) for a longer time than it is for a typical transaction. That is a very NARROW vulnerability (not that it is justified ...). The paper clearly says on a large set of ATMs they could NOT decipher the "algo" for the UN generation. This is a exploitation of a very very corner case.
The paper also clearly says that EMVCo HAS ALREADY published rigorous tests to test the randomness of UN generation (before this paper was published).
So the title here, in the BBC website and some of the comments are way off. (understand that BBC and /. have to have readership ...) Couple of additional comments, EMV cards are unclonable (so are the SIM cards used in phones which use similar technology), the standards are open (you can download the standards for free from the emvco website) and there are plenty of fraud detection algos running on issuer servers to detect suspicious transactions. The paper in the second page unambiguously states that AFTER the introduction of EMV cards "card-not-present" transaction fraud went up, precisely because EMV cards are secure.
There will be always studies like this which exposes flaws (this particular one was an extremely corner case) which generally strengthen the current systems. I have followed the research coming out of cambridge on related topics (have exchanged notes with some of them), they are fine researchers and if you read the paper, you will see that they are NOT saying EMV is insecure but are identifying corner cases and defective implementations.
Cheers,
-Bhaktha
Re-read your chip & PIN liability statements. Chargebacks with chip & PIN are very difficult to do and weighed heavily against the cardholder.
By default, if a transaction is conducted via chip & PIN, the consumer is liable for all charges. The use of a PIN constitutes, in the eye of the bank, de-facto shift of liability for the transaction. In the event of a dispute, it is up to THE CONSUMER to provide evidince that he / she did not perform the transaction. This is a marked shift from the old magstripe / signature liability, where it was up to the merchant to prove that it was you making the purchase in a dispute. Now, it is up to the consumer to prove it WASN'T you - good luck with that!
I am glad people are finally waking up to this because I avoided chip & PIN as long as possible due to this, but it is being rammed down our throats, along with this liability shift, and no one is noticing.
http://www.youtube.com/watch?v=B80SyRmtbdI
I know it happened 12 years ago, but come on, the chip cards with pin have been cracked and crackable for a long time. In 2000, Serge Humpich, a french hacker found a flaw in the chip design and used Japanese algorithm to factorize the prime used in the chip card.
In French:
https://fr.wikipedia.org/wiki/Serge_Humpich
http://www.bibmath.net/crypto/moderne/cb.php3
In English:
http://www.theregister.co.uk/2000/02/26/french_credit_card_hacker_convicted/
http://www.amazon.com/Serge-Humpich/e/B001K7H3DE
I remember my reaction when chip cards appeared in Canada *after* 2000, as if they were waiting on having a backdoor before they deployed them.
JigJag
"The hallmark of humanity is the ability to move beyond sensory inputs" - Mary Helen Immordino-Yang
The main problem with chip-and-pin, from the consumer's perspective, is that it shifts the liability onto the CARDHOLDER, not the merchant. The issuers insist that merchants bear the liability for old magstripe transactions, but for chip-and-pin transactions it is presumed that you, the CARDHOLDER, are responsible unless you can *prove* otherwise. That's why the merchants were all so eager to get the chip-and-pin hardware deployed... it reduces their fraud costs (shifting them onto the victim cardholders instead).
Here's this attack in a nutshell:
The protocol between card and ATM incorporates an "Unpredictable Number" which is generated by the ATM and sent to the card as part of a transaction request. The card returns a response which includes this Unpredictable Number, and is encrypted with a secret symmetric key stored on the card. The other copy of the symmetric key is known only to the issuing bank. The ATM sends this response to the issuing bank over the network, where the transaction is vetted and approved.
The important role played here by the "Unpredictable Number", is to guarantee the _freshness_ of the transaction to the issuing bank: its how they know that the challenge sent to the card, and the response returned from the card, were generated _while the user was using the ATM_ and not at some much earlier time. Unfortunately, the party relying on the unpredictability of the number is the issuing bank (the one who issued you the card) and the party *generating* the number is the ATM, which might be in a different country, might be operated by an adversary, might be compromised by malware, might be in a Mafia-owned store and have been tampered with, etc. To be secure, the number should have been generated by the issuing bank at the start of the transaction, but the system is not designed that way (probably because it would slow the transactions down too much). So instead of a few hundred issuing banks, you're relying on literally _thousands_ of different ATM manufacturers and operators, to securely generate unpredictable random numbers for you. But many of them don't... they use crappy generators like stdlib rand() or system timers which can be forced into a known state by power-cycling the ATM.
If the attackers can predict what "Unpredictable Number" the ATM will generate (and using the techniques from the paper, they often can) then that means they can send those numbers to the user's card when its inserted in a compromised ATM or POS terminal, and get the card to encrypt their illicit "request" as needed. Then at some later time (maybe days or even weeks later) they present the card's response to a real ATM somewhere else, and take money out of the cardholder's account. The attackers have to choose the amount and the date of the attack in advance, but they can use any vulnerable ATM in the same country as the compromised terminal where the cardholder's info was skimmed from.
So this attack is basically as strong as card cloning. There's basically nothing you could do with a cloned card, that you can't do with this attack.
That's right, this is at least the second independent way Chip & Pin has been found to be broken. The banks claim to have multiple layers of security, but what they actually have are multiple breaches of security.