That means the retailers who spent more to acquire this particular flavor of two-factor authentication may actually be less secure right now than if they relied only on strong passwords. After all, two-factor security effectively becomes one-factor security as soon as users start trusting too much in one factor (such as SecurID) and get sloppy with the other (their regular passwords). That's a natural danger—after all, how many people use two different keys to unlock their front door?
RSA's shift came after defense contractors Lockheed Martin and L-3 Communications both were reportedly hit last month by attacks using information from the SecurID system that was stolen from RSA in March, along with "social engineering" used to collect conventional password information.
To calm customers, RSA Executive Chairman Art Coviello said the company will offer replacement fobs to "customers with concentrated user bases typically focused on protecting intellectual property and corporate networks," and also offer "risk-based authentication strategies for consumer-focused customers with a large, dispersed user base, typically focused on protecting Web-based financial transactions."
Right—that's certainly vague enough to include or exclude almost any retailer who is a SecurID customer. Do retail networks that handle payment-card transactions qualify as systems that protect intellectual property? If so, those users get new SecurID fobs. Or do retail chains count as "consumer-focused customers with a large, dispersed user base"? In that case, they just get security advice.
And retailers already know what that security advice is likely to be: Make sure users guard their passwords as closely as they guard their SecurID fobs. And that's not easy to accomplish. Give users—even highly technical users working in IT—a shiny new high-tech device, and they're likely to feel it's much better than just an ordinary password. Just by itself, that's true.
But the next, nearly inevitable conclusion is that the high-tech device renders the ordinary password obsolete. If the result is that users no longer worry about passwords—who cares if it's weak or someone finds out? I've got a SecurID fob to protect me!—that multi-factor authentication system becomes single-factor. And if it's cracked (as all security technology eventually is), that can drop suddenly to zero-factor.
It's not a problem limited to SecurID users, either. When Google announced its forthcoming digital wallet last month, Google Payments VP Osama Bedier and other executives went on at length about the fact that Google Wallet stored payment-card numbers in a Secure Element that was encrypted and couldn't be cracked even with a laser. (That's less impressive to anyone who knows that all systems using near field communication hardware have Secure Elements meeting that description; there's nothing special about Google's version.)
Then Bedier performed a live demonstration that showed the screen of his phone—which ended with Bedier's personal payment-card information being displayed on giant screens for his audience. That laser-proof encryption didn't help the self-professed "security freak." In fact, if he hadn't been lulled into a misguided sense of security by that impressive-sounding technology, he probably would have been more secretive with his payment-card information.
That suggests the best security advice retailers who use SecurID can implement: Make sure your users are very worried about their passwords. Until your chain moves high enough up the list to have your fobs replaced (or your lawyers rattle RSA's cage hard enough), passwords are the most secure thing standing between cyberthieves and your systems.
And once those SecurID fobs are replaced, those passwords are still the only thing sure to keep that expensive two-factor authentication system secure.