IE 11 is not supported. For an optimal experience visit our site on another browser.

RSA’s silence spreads fear about use of SecurID

Four days after security firm RSA announced that it had been hacked, and then refused to say more, experts were openly questioning whether any of its 40 million SecurID tokens should continue to be used.
/ Source: SecurityNewsDaily

Four days after security firm RSA announced that it had been hacked, and then refused to say more, experts were openly questioning whether any of its 40 million SecurID tokens should continue to be used.

“Any company using RSA SecurID tokens should consider them completely compromised and should insist upon their immediate replacement,” independent security specialist Steve Gibson posted on his personal blog.

“I’m speculating, but I’m pretty confident that somebody has the root seed file,” an anonymous former RSA employee told the New York Times, referring to the master list of unique numbers assigned to each token.

That secret 16-digit “seed” number is combined with the time, a hash algorithm (also technically secret) and possibly the token’s serial number to create six- or eight-digit passcodes, which change every 30 or 60 seconds.

Each employee of an estimated 25,000 corporations and government agencies around the world needs to type in that passcode, along with a username and personal password, in order to remotely access his or her organization’s internal network.

At the other end, an “ACE” server uses a synchronized clock and the same hash algorithm to decrypt the passcode. If the result matches the seed number of the token assigned to the user who is logging in, the user is accepted.

It’s a pretty simple process, but it absolutely depends on the seed numbers being known only to the ACE server and to RSA. (The end user does not know the seed.)

A similar process is used by wireless carriers to authenticate the identities of 250 million smartphones worldwide.

Were an intruder to get a list matching serial numbers (which are printed on each token, and easily discoverable on smartphones) with seeds — the “keys to the kingdom” — then wireless, corporate, governmental and even military security would be severely compromised.

Each network would be protected only by a username and password, which are relatively easy to crack using computers that rapidly spin through possible combinations.

If that were the case, every single RSA token in existence would be useless and would need to be replaced — the key fob-sized gizmos are not upgradable. (It’s possible that a software upgrade could remedy the smartphone weakness.)

Unfortunately, RSA’s vague statements to clients and to the Securities and Exchange Commission on Thursday, and the company’s lack of subsequent statements, have done little to dispel speculation that that’s exactly what’s happened.

Dell’s SecureWorks blog, while more polite than some other warnings, was blunt.

“Until further information is available,” it said, “the prudent course of action is to assume the worst: that SecurID seeds have been exposed, their assignment to specific RSA customers is known, and the source code of ACE server and other products has been compromised and may reveal weaknesses.”