Security researchers have discovered a flaw affecting Google's Chrome browser that exposes it to "clickjacking"--in which an attacker hijacks a browser's functions by substituting a legitimate link with one of the attacker's choice.
Google has acknowledged the flaw and is working toward a patch for Chrome versions 188.8.131.52 and earlier when running within Windows XP SP2 systems, according to SecNiche security researcher Aditya Sood.
Sood disclosed the flaw on Tuesday and has since posted a proof of concept on the Bugtraq vulnerability disclosure forum.
"Attackers can trick users into performing actions which the users never intended to do and there is no way of tracing such actions later, as the user was genuinely authenticated on the other page," Sood said within the disclosure.
While Google is working on a fix, a representative for the Australian arm of the company pointed out thatcan affect all browsers, not just Chrome.
"The (clickjacking) issue is tied to the way the Web and Web pages were designed to work, and there is no simple fix for any particular browser. We are working with other stakeholders to come up with a standardized long-term mitigation approach," they said.
However, Nishad Herath, an independent security researcher and CEO of Australian security consultancy Novologica, told ZDNet.com.au that after running Sood's proof of concept he found that Internet Explorer 8 (release candidate 1 and beta 2 versions) and Opera 9.63 (the latest version) were not exposed to the flaw. But, like Chrome, Firefox 3.0.5 was exposed.
Google's security researchers had not found any attacks in the wild that exploited the specific vulnerability, said Google's representative.
"Clickjacking means that any interaction you have with a Web site you're on, for example like clicking on a link, may not do what you expect it to do," explained Herath.
"You may click on a link that looks like it's pointing to a picture on Flickr, but in reality, it might first direct you to a drive-by-download server that serves malware. These types of attacks can be used to make you interact with Web services you're already logged onto in ways that you would never want to, without you even knowing that it has happened."
Liam Tung reports for ZDNet Australia from Sydney.