Businesses are growing more concerned about the use of social networks, starting with Facebook.
According to a survey of 502 IT professionals by Sophos, businesses are seeing more malware and spam, and 60 percent of respondents put Facebook ahead of MySpace, Twitter and LinkedIn as the riskiest social networking site. The statistics, which were included in Sophos' "Security Threat Report: 2010" (PDF), revealed that while 33 percent block Facebook for productivity reasons, businesses are also concerned with the prospect of spam, malware and data leakage on social networks.
"Furthermore, over 72 percent of firms believe that employees' behavior on social networking sites could endanger their business's security," according to the report. "This has increased from 66 percent in the previous study (in April). The number of businesses that were targets for spam, phishing and malware via social networking sites increased dramatically, with spam showing the sharpest rise from 33.4 percent in April to 57 percent in December. This highlights a surge in exploitation of such sites by spammers."
While just 21 percent of the respondents in the April survey said they or their colleagues had received malware via a social networking site, that percentage increased to more than a third in December. When it comes to Facebook in particular, 45 percent of respondents said they do not control access to the site. However, Sophos Senior Technology Consultant Graham Cluley said seven percent did so due to fears of data leaks, while 11 percent controlled access because of malware concerns.
"There have been a variety of malware attacks via social networks - but Koobface is the Godfather," Cluley said. "There have been many different versions of Koobface, and it has become steadily more sophisticated - attacking a wider variety of social networks and becoming much cleverer in the way that it operates."
Users of social networks can also face a more indirect risk - attackers using the sites to conduct surveillance on potential victims and ultimately compromise them. A spokesman for McAfee said this method is believed to have been used in recent cyber-attacks, with attackers moving "a degree away from their ultimate target and hitting someone in their circle of friends...(before) moving in on the target assuming the identity of a friend of colleague."
"Undoubtedly a large part of the incentive of social networking attacks is to compromise the victim's machine and infect it with malware that turns it into part of a bot," Cluley said. "As such it can be spied upon (and personal information stolen) as well as abused to send spam/spread malware etc."
As such, compromised accounts can have real value on the black market. According to Dmitry Bestuzhev, senior regional researcher for Latin America at Kaspersky Lab, said recently a Twitter account was seen being offered for $1,000 on a hacker forum. In the hands of cyber-criminals, the accounts can be used to spam out malicious links that lead to malware infections, he said.
"Many tweets today come with URLs and not just clear text URLs, but short URLs," he said. "Really few followers check first what's behind each link, they trust the content and they click any link in a tweet. This is a perfect opportunity for the criminals to infect more people, doing it fast and without any suspicion."
The more followers a person has, the more potential victims there are, he added.
"If it's a Trojan.Banker or Trojan.PSW or Trojan.Spy family, any and all money-related passwords may be stolen and of course, it becomes money for the criminals," he said. "Of course, if they can now steal another Twitter account they can infect more and more people. In other words, it's a kind of pyramid- they would infect followers of an infected follower of the initial Twitter accounts."