Take the 2-minute tour ×
Information Security Stack Exchange is a question and answer site for Information security professionals. It's 100% free, no registration required.

Picture an ATM security system where, if the user types the PIN in reverse, the ATM alerts police, the customer's bank, or other parties.

Are there any ATM manufacturers that use this or a similar system? Is it secure?

share|improve this question
7  
What happens when my PIN number is 1221? –  Hoang Minh Jan 16 '13 at 7:54
    
I used the principle "add 1 to the PIN" instead in one application, i.e. 1222, and don't issue 9999. –  Callum Wilson Jan 16 '13 at 8:56
    
Reminds me of the FAA policy that if you're ever hijacked, you set your transponder code to 7500. Government representatives in intimidating body armor will then meet you at your destination with guns drawn. This rule ends up being primarily a source of funny stories swapped between ATC crew. –  tylerl Jan 16 '13 at 10:00
3  
@Hoang Minh In US patent 5731575, which covers this technique, the scheme includes a requirement that you don't allow customers to choose a palindromic PIN. –  Graham Hill Jan 16 '13 at 14:25

6 Answers 6

This type of senario is quite common in other applications. While I have never heard of it being implemented on an ATM, I don't see why I couldn't. This is often implemented in home security systems. If someone has you at gun point, you enter a "dummy pin" and the system disarms but a call is made to the authorities.

share|improve this answer
1  
If the reverse system were ever implemented, then palindrome pins wouldn't be allowable. Duress codes are indeed a reality for alarm panels, but most ATMs don't have one, first because most users have trouble with just one PIN they use regularly, and second because ATM transactions are supposed to be fast, so entering a duress code would still give the BG the money they wanted long before a police officer could show up. A home invasion or store robbery takes more time for the BG to get the goods, giving police time to respond. –  KeithS Jan 16 '13 at 19:27

This is an urban myth. The FTC looked into the whole thing in 2010 and could find no bank or manufacturer of ATMS in the US who had ever implemented such a system; they also concluded that it would be expensive to implement and probably not work very well.

Duress codes do exist in other areas - some home alarm systems have them, either in the PIN or in the code used over the phone to the alarm company to cancel an alarm.

There's an interesting paper from the University of Waterloo that points out some of the issues with such systems. For example, because of Kerchoff's principle, we must assume the bad guy knows there is a duress code, so she can demand the victim tell her that as well and improve her chances to 50%. And of course there's the old chestnut of attacking the password change process.

share|improve this answer
1  
I work for an alarm/surveillance company, and some of our systems are even set up with two duress codes for exactly the situation the Waterloo paper describes; why give the BG a disarm code and a duress code and give him a 50% chance to enter the disarm code, when you can give him two duress codes and thus a 0% chance? The criminal would have to be a true insider to know you're running a shell game on him, and if he's a true insider there's really no good way to stop him, but plenty of other avenues to catch him. –  KeithS Jan 16 '13 at 19:09
    
Even with only one duress code, your bad guy would have to be in the upper 1% of the sophistication bell curve not to fall for it. Systems (both electronic and human-guided) are designed to make the duress code appear to be the disarm code, so doors unlock, keypad reports disarmed, no sirens, no flashing lights. The agent is usually instructed not to interact with the site at all in a duress, but in some circumstances will further the illusion with their "normal" opening script. The BG has no clue it's about to go badly for him until the entire local precinct shows up outside the building. –  KeithS Jan 16 '13 at 19:17
    
@KeithS Ah-ha, you've found Auguste's secret weakness! Good threat modelling can sometimes trump the Kerkhoff principle. –  Graham Hill Jan 17 '13 at 11:07

I have another idea in that issue, there may be two different pin, the first one is normal pin, second one is backup pin, when you are in danger, you enter that pin and it will show the amount of money you have as, let say, 10% of your money.

Consider you have 1000 usd in your account, somebody put a gun on your head, and you entered that pin and you lost 100 usd, and you are most likely be saved with 100 usd instead of 1000usd.

I dont know weaknesses of that story by the way.

share|improve this answer
    
Interesting, but I would imagine this may be too complicated for most people, considering how many PIN resets are requested. –  Rory Alsop Jan 18 '13 at 13:20
    
@RoryAlsop, Yes, but this is usability vs. security problem. Not every technology is used by everyone –  tusherity Jan 19 '13 at 8:52
    
ATMs, though, need to be used by anyone, so the requirement is very different to an alarm system for example. –  Rory Alsop Jan 19 '13 at 8:55
    
Novice users have one pin, others will have two pin. If 10% of users are safer with this technology, this will be a success, I guess. –  tusherity Jan 20 '13 at 10:10

If this were implemented, then if someone was going to rob you at an ATM, they would just wait for you to enter your PIN, then attack you afterwords. Since there are key-tones, its not like they cant hear you entering in your code, THEN grab you.

share|improve this answer

Actually never heard of this before. If this is possible it would be a good security measure to follow, but the problem is that it will not work with palindrome pins.

share|improve this answer
    
This is more along the lines of a discussion; SEC:SE should be closer to a Q&A site. Your last line makes a good point though –  Mark C. Wallace Jan 24 '13 at 12:11

i think there's a drawback in each implementation.. for instance tusherity's response clicked a lot to me. the idea of two pins, one of which shows an incorrect balance.. but if i, as a theif, know that a particular bank offers two pins, and the guy enters his pins to show me a balance of $100.. id double check that by forcing him to enter the other pin.. if the account now shows $10, id know he was telling the truth.. if he hesitates, or if the account indeed shows $1000, id know he was lying.. unless he's really smart and can convince me that he hasnt opted for a duress pin.. i just need to have an account at the said bank to know they DO offer duress pins..

in my opinion, when it is time for you to get robbed, you will be robbed, no matter how elaborate a duress scheme is in place.. you can try staying a step ahead of the criminals, but they soon catch up, always.. they adapt to the environment..

also, as pointed out in previous answers, the authorities would almost never be on time.. in my country, a lot of armed robberies take place.. a guy goes into the ATM room, forgets to lock the door, is drawing out cash and two armed men enter, rob him at gunpoint and flee in seconds.. in my country, the best rules are these..

  • dont venture near ATMs when its deserted nearby or suspicious ppl are nearby..
  • if you are cornered, let them have whatever they want. they will not hesitate shooting you or seriously injuring you.
share|improve this answer
    
"I never use the duress PIN so I forgot it." Most ATM robbers aren't going to shoot somebody because they forgot their PIN, especially if they've provided at least one PIN. –  Tom Marthenal Jan 26 '13 at 4:44

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.