AOL has patched a serious flaw in its instant-messaging software, but more troubles may lie ahead, according to a security researcher familiar with the issue.
The flaw affects how the AIM software uses Internet Explorer's software to render HTML messages. By sending a maliciously encoded HTML message to an AIM user, an attacker could run unauthorised software on a victim's computer or force the IE browser to visit a maliciously encoded Web page.
Although AOL says it knows of no attacks that exploit this problem, security experts like Aviv Raff had warned that the flaw could possibly be used by a self-replicating computer worm attack.
AOL was first informed of the problem around the beginning of September, and the 6.5 update, released on October 3, was supposed to address this bug. However, Raff said he'd examined the AIM 6.5 release and discovered that AOL had not fixed the underlying problem.
"I've tested this version against the critical vulnerability I've found. While it does fix the specific attack vector of the vulnerability, it still does not utilise the Local Zone lock-down," he wrote in a blog posting Sunday. This means that an attacker who discovered some new way to insert malicious script into an HTML AIM message could end up running unauthorised software on the victim's machine, he explained.
Raff had planned to release proof of concept code showing how this flaw could be exploited by an attacker. But now he says he's holding off on that disclosure until AOL "properly" fixes its AIM client. He believes that with a little bit of tweaking, criminals could probably adjust his proof of concept to get around AOL's protections and create a "massive IM worm."
AOL representatives were not immediately available for comment.
The problems that AOL is experiencing with its messaging client are not unlike those Microsoft has already run into with the browser, said David Marcus, security research and communications manager with McAfee's Avert Labs. As more features are added, they sometimes create new ways for hackers to attack. "When you're adding all this interactively and the ability to run this HTML code through your client, you're inheriting malicious issues," he said.