AOL Moves Beyond Single Passwords for Log-Ons

ars writes "Yahoo is reporting that AOL is adding a new feature alowing customers to use two passwords to log on. The second password comes from a small small device from RSA Securitywhich displays a new password each minute. The scheme is called two-factor authentication and will cost $1.95 a month plus a one-time $9.95 fee. It's aimed at small business and people who conduct large transactions online."

Comments

This has been used internally for years

[David_W]

Interesting... this particular feature has actually been a part of AOL for several years now. All AOL employees are issued SecureIDs and are required to use them to log in to various places. It seems they've just expanded the feature to non-employees.

Re:This has been used internally for years

[LetterJ]

A lot of companies use them for their VPN access. Several of the last big companies I've contracted for have required them. Some just use the value from the fob and others require a concatenation of the fob value and a prechosen password.

Unfortunately, I've found that the fobs tend not to enjoy the abuse that being on my keychain tends to bring. The LCD panels end up pretty scratched by the time I'm done with them.

Re:This has been used internally for years

[LetterJ]

I haven't had a battery go dead in one yet. Granted, I haven't had the same one for longer than a year, but physically, the display is pretty much what a digital watch would be. There's no backlight, etc., just a string of numbers and a little countdown meter. Internally, it's doing more calculations than a watch does, but we're still talking about a really small electrical draw.

Incidentally, there's an expiration date on the back of these things (I just thought to check). My current fob has an expiration date in Dec of 2007. I think that's a pretty good duration and it's more likely the thing will get destroyed by being dropped on the pavement, lost, scratched beyond usability, etc. in over 3 years of use on a keychain.

whoo.

[nbvb]

SecureID.

Whoo.

Been there, done that.

All it does is make an attack "more" difficult, but nowhere near impossible:

http://www.tux.org/pub/security/secnet/papers/se cu reid.pdf

Re:whoo.

[Fedallah]

After reading through the paper, I have to say that the attacks contained therein are simply not that impressive. In it, the author describes the following attacks:

• An race attack that is only valid if the user slowly logs in over an unencrypted non-line-buffered telnet session using the SecureID. I have never seen an implementation of SecureID used like this, and we can be assured AOL's implementation will not be susceptible (as they will undoubtedly be having the token typed into a local window, not transfered over a network character-by-character)
• A attacked on a clustered implementation where the attacker shuts down several lines of communication as part of the attack. This is probably the closest thing to a dangerous attack; however, the author even describes a way that the servers could be programmed as to avoid this situation. At the time of the article, this has not been implemented in the server, but apparently, the article was written in 1997 (or thereabouts)
• A software bug in an older version of the software. Shameful, yes, but apparently fixed about 8 years ago.
• A theoretical attack of which the author claims "It is not known whether all of the semantics are
  absolutely correct in this example but it is quite probable that some variation of the
  attack is possible."

Of course, I'm not claiming that the security of a SecureID implementation is unassailable, or that SecureID is a panacea for security problems. I just don't believe an old article that describes some irrelevant not-quite-attacks is sufficient to cast doubt on the extra security provided by SecureID, and that attacks on SecureID are actually much more difficult than you seem to be claiming.

Useless

[cly]

When common folk's computer is still infested with adware/trojan/god-knows-what

This just creates an illusion of security.

Re:AOL Security at work again...

[Anonymous Coward]

RTFA you nincompoop... one of the passwords changes every minute, and it's generated automatically. So phishing attempts would not be all that successful.

Re:This will make the problem disappear.

[JohnHegarty]

two points...

1) it only lasts 60 seconds
2) if used , it can't be used again until the minute is up

Re:Not a bad idea

[PugMajere]

When they go out of sync, either they haven't been used in a *long* time, or the server's clock is drifting badly.

The server is designed to track slight drifts in time and track/compensate for the cards.

Even if they are out of sync, the most you have to do is enter two codes instead of just one.

I Used AOL securID

[Apple Acolyte]

In addition to being used internally by AOL, securID was offered to some regular users who were targeted by hackers. Like an organization I work for. The securID token is smaller than the average pager, having no buttons, only a display with a string of numbers that would alternate every 30 seconds or so. The biggest shortcoming of the system is that the battery did eventually die, and there was no easy way to replace it. That meant the account in question had to be unbound from the token. And it took a long time to find a rep that could actually handle that request. (Not that that was too big of a deal, since my organization only kept its AOL account alive for legacy purposes.) In terms of use, however, the token was not obtrusive at all. No additional client software was required. Upon sign on, a securID window was presented prompting the user for the key. Otherwise, it was transparent.

The big question is, is AOL's true motivation for offering this to regular customers just to compensate for the service's renowned terrible security?

Re:Good deal - basic math?

[Meostro]

How exactly does $9.95 plus $1.95 per month get to be $60/yr?

1.95 * 12 = 23.4

23.4 + 9.95 = 33.35

33.35 != 60

Re:Time Drift - sliding window

[morzel]

IIRC RSA uses a sliding window to correct for time drift.

In an ideal world, the server and the fob are perfectly synchronized, meaning that the server knows which number the fob will generate at any given time. In the real world, the fob creeps behind/before schedule and generate a number x entries before/after the expected entry.
If this is the case, the server looks up if number x is in the vicinity (e.g.: within 5 minutes) of the expected number. If that's the case, the server assumes that the clock has drifted and marks the amount of time that the fob has drifted for next authentications.
If x is outside that range, but inside a much broader range (e.g.: one hour), it will request the number that the fob generates next, and checks if that number matches the one that should come after x. Then it marks the drift amount and allows access.

The server automatically compensates for inaccurate clocks in the fobs; as long as you use it regularly. Only if you have,'t used your fob for quite some time, and it has a really lousy clock they de-synchronize, requiring a hardware swap (and/or manual intervention from the sysadmin).

Aol must really care about security...

[SirTwitchALot]

because they can't be making much money from this:

RSA sells these devices for $60 each or so in bulk. RSA fobs are programed to expire in 36 months. Let's say AOL got them for $50. The customers are paying 9.95+(1.95*36) or $80.15 over three years. That gives AOL $30.15 or about $10 a year. I'm sure aol could find some other way to fleece their users less than a dollar a month, leading me to believe this isn't just some profit making venture (not to mention the cost of the servers to implement this, which is not insignifigant.)

Got a good screen name? Get one of these.

[YetAnotherName]

If you're lucky enough to have a decent screen name on AOL, like your first or last name, then you probably want to get one of these devices.

When I got my Yahoo account years and years ago I was early enough to get decent screen name. The problem is that today that account is routinely hacked (and once, even pwned, but thanks to the nice security folks at Yahoo, given back to me). People don't like to use something like "%geeba%56672" for Yahoo Instant Messenger. I imagine the same thing is true on AOL. Having a smartID or securiCard or other defense would be nice.

(Then again, auctioning off a nice AOL screen name might be worth a few bucks on eBay...)

But you don't need "two" passwords !

[syrinje]

Two factor authentication relies on (d'uh) two inputs to the authentication algorithm - something you know (like your username) and something you have (like a password - whether generated by a SecurId or not).

The advantage of the automagically generated password is that the password is a temporal function of the account. This means that the server and the password generator both work off the same clock base to calculate a password for your account and authentication succeeds if the two match (within some non-zero time window - to compensate for clock drift). the password is thus valid for a very short duration and makes it very hard for a MIM to capture, replay and use

As far as I can see the first (user memorised password) is merely an artefact of an older system left in there to make the user feel good about having some password control since that is the fator that is most vulnerable to compromise (think social engineering).

A more robust mechanism would be to add a challenge response to this mechanism - the suthenticating system gives you two numbers (n1, n2)which you feed into your password generator and it generates the response thus -

R sub t = f(t, n1, n2)

The authenticating system performs the same computation and accepts your password if it matches with the result generated locally. Banks in Sweden have been using this for quite a while now - the password generator is, of course, protected by a PIN number to unlock it for use and therin lies the weakest link!

Re:But you don't need "two" passwords !

[Gunzour]

Authentication can generally be done using any combination of these 3 factors:

- Something You Know. Generally a shared secret, such as a password.

- Something You Have. Prove that you are in possession of something. By entering the code from a SecureID card, you prove you are in possession of the card. A physical key entered into a lock is also Something You Have. The CVV code on the back of a credit card is a weak form of Something You Have (it could be argued it is something you know, but online stores are using it to 'prove' you are in possession of the card).

- Something You Are. This is biometric authentication, such as voiceprint, fingerprint, iris scan, DNA check, dental records, etc.

Your username is only a bit data -- well-known data at that. It doesn't count for any of the three factors.

Re:Synchronized Clocks?

[PalmerEldritch42]

The server does allow a range of codes to work. I have been using SecurID and you can put in the tokencode from 1-2 minutes ago and it will let you in. So, if the token gets out of sync from the server, it is ok. If it gets too out of sync, then you need to call the help desk and they can resync it using some online tools. It takes less than a minute to do. I've never experienced a time drift problem that resyncing didn't fix, but theoretically, if it cant sync back up, they can always just send you a new card and use that one instead.