Social Engineering: The Good Guys Strike Back
If you can't beat them, scam them back...or slander them into quitting.
That's the approach some researchers at Carnegie Mellon University are suggesting for making it harder for individuals to trade in malware and stolen financial and identity data in the Internet black market.
A lot of the illegal activity that is happening on the Internet these days is readily accessible to absolute newbies as well as to experienced professionals, said Jason Franklin, a doctoral student at CMU's computer science department.
"What used to be a true underground market has emerged more publicly," Franklin said. "It's very easy for anyone to identify forums, chat rooms and other locations where people are trading illicit goods and services of all sorts."
One way to disrupt this booming economy is to make it more unreliable and costly to participate in such transactions, he said, especially for newcomers. The idea is to use slander attacks and other techniques aimed at undermining the verification and reputation system used by cybercrooks, he said.
The suggestion is based on a seven-month study of one underground site by Franklin and three other researchers -- one from CMU, one from the University of California, San Diego, and another from the International Computer Science Institute. The purpose of the study was to measure and quantify the scope of the illegal activity that was happening on such sites.
During that period, the researchers counted more than 80,000 stolen credit cards and illicit goods worth an estimated US$37 million offered up for sale on the site.
Honor among thieves, sort of
Buyers interested in purchasing such items typically contacted the seller using e-mail or private instant messages, and transactions were paid for using non-bank payment services such as e-Gold. "These markets have a system for assessing how reliable a buyer or a seller is," Franklin said, explaining that trusted third parties that provide a "verified" identity status to buyers or sellers who have established a track record for keeping their end of the bargain in an underground transaction.
In a transaction between a verified seller and an unverified buyer, the buyer pays upfront for the item being transacted before actually receiving it. "It's just a convention," says Franklin. Conversely, he said, "the unverified seller will give you the credit card numbers before you provide payment because there's risk involved."
Typically, buyers and sellers are conferred 'verified' status by the operators of the IRC channel in which they are doing business and are identified by little voice administrator flags against their names. To earn the status, a brand new seller may sometimes distribute stolen card numbers for free to others on the IRC channel to demonstrate his access to such information, Franklin said.
One way to disrupt this setup is to create a deceptive sales environment using a so-called Sybil attack, Franklin said.
A Sybil attack is a way of subverting a reputation system by overwhelming it with numerous forged identities or Sybils. "Since these markets have emerged from the underground, they are allowing anyone" to participate in them, he said. "It is basically trivial to connect multiple times and create multiple IDs," said Franklin, whose team also developed a technique to establish fake verified-status identities that are difficult to distinguish from other verified-status sellers.
The goal is to make it hard for buyers to identify the real verified-status sellers from the fake ones. Using the technique, it is conceivable for someone -- from law enforcement, for instance -- to engage in a transaction as a verified buyer or seller and then not fulfill his side of the deal. Doing this often enough could undermine the credibility of all verified-status identities, especially for newcomers, Franklin said. "This won't disrupt already established relationships, but it will make it harder for newcomers" to establish trust relationships, he said, adding that there may be legal implications that need to be explored before such an attack can be implemented.
Another approach suggested by Franklin and his team are slander attacks against those with genuine verified-status identities. This tactic takes advantage of the primitive processes most illegitimate IRC channels use for handling complaints of false transactions, allowing slanderers to wrongly defame someone. Such defamation could easily be accomplished using Sybils with verified-status identities.