Social sites' insecurity increasingly worrisome
'Fertile ground for malicious coders'
Personal web spaces on MySpace, videos on YouTube, and blogs - community sites hosting user-created content - have become increasingly popular.
While the web has always been about publishing digital information, the stunning popularity of hubs for content created by the audience has attracted more people to the world of quick-and-easy publishing, but the trend has some security experts worried.
In November, security firm Websense alerted internet users over a handful of MySpace pages hosting videos that, when played, attempted to install adware on a viewer's system. The videos used the digital rights management facilities built into Windows Media player to start installing the software, earning the fraudster money as an affiliate of adware purveyor Zango.
The incident underscores that such content should not be trusted, said Dan Hubbard, senior director for security and technology research at Websense. As more internet companies develop tools for turning their audience into the prime source of content, online fraudsters and data thieves are looking to exploit the systems to reach mainstream audiences, he said.
"User created content is definitely a big security shift," Hubbard said. "I don't even think the companies have really thought about how to control things that they don't have (direct) control over."
The number of incidents involving user-created content hubs is increasing. Microsoft researchers have found that a loose collection of websites, or an "exploit net", play host to malicious content and use comment spam to attract potential victims. And social networking sites are at the centre of the storm. For example, a large number of the intermediary sites, as many as 17,000, are hosted on Google's Blogger service.
The internet search giant has its eye set on services that turn visitors into content creators. With Google's $1.6bn purchase of YouTube, the popularity of user-created content hubs will only rise. Giving the audience the tools to turn their creative energies into attractive content is a key piece of that popularity puzzle, but the sites need to weigh such decisions against the security implications, said Christopher Boyd, director of malware research at messaging security firm FaceTime Communications.
"It's a huge problem," Boyd said in an email interview with SecurityFocus. "These sites rely on an anything goes approach to attract users, with pretty much everything you could think of switched on for the user to customise."
And that makes the sites a potentially fertile ground for malicious coders and online fraudsters, he said.
MySpace failed to comment on the issues after being contacted numerous times.
It's not just MySpace that finds itself the focus of fraudsters, however.
Wikipedia, the online community encyclopedia, has also had to deal with such problems. The various Wikipedia sites that allow online users to add and edit content could open the door to potential malicious content, according to security experts.
That's almost what happened in November when a fake site masqueraded as a German version of Wikipedia hosting an entry on a variant of the MSBlast, or Blaster, computer worm. Instead, the web page attempted to compromise visitors' machines. While neither the site nor the content had been hosted on Wikimedia's servers, the phishing scam had such polish that it fooled at least one antivirus firm. The Wikimedia Foundation, the organisation that operates the Wikipedia sites, has seen the writing on the wall and taken steps to limit what users can do.
"We do not allow linking from executables," said Brad Patrick, general counsel for the Wikimedia Foundation. "The intent here is that we are always providing our readers with at least one additional step between us and any malicious content."
Even the virtual world of Second Life, which depends on user-created content to keep its economy going, has had to deal with virtual viruses, known as "grey goo."
For about about two hours on 19 November, the company that manages Second Life, Linden Lab, scrambled to contain an outbreak of Sonic the Hedgehog-esque gold rings. The objects spread within a region, slowing down the servers that maintain the Second Life world, or grid. It's the third major attack since September, each time the world has been overrun with quickly reproducing digital objects that have taken hours to clear out of the system.
Google, Wikimedia, and Linden Lab have all built defenses into their systems, and MySpace has hired former Microsoft investigator Hemanshu Nigam to beef up the social networking site's security.
"Specific to our own products and properties, including sites which host user generated content, we work constantly to prevent people from misusing our services to distribute malicious software," Barry Schnitt, spokesman for Google, stated in an email interview. "When we become aware of an instance where this happens, we take immediate action to limit user exposure."
That's good, but the companies need to attack the broad range of threats, rather than focus on, for example, child porn at the expense of malicious videos, said Websense's Hubbard.
"A lot of their security is geared towards child pornography, taking down content that they don't want on their site," Hubbard said. "They need to get more savvy and build up their security teams, because we are talking about hundreds of millions of pages changing daily."
And, at that rate, the risk will only likely increase, he said.
This article originally appeared in Security Focus.
Copyright © 2006, SecurityFocus