Meta does the 'We found baddies and crushed them' thing again – this time for AI

Who would have thought crims would try using Facebook to fool people?

Meta says it has shut down over 1,000 links related to ChatGPT that lead its users to malware, as criminals seek to profit from the current craze for generative AI.

ChatGPT has quickly bagged more than 100 million users, encouraging many organizations to explore how generative AI might help them increase productivity and profit. Scammers are thinking along the same lines, offering links and other stuff related to the chat bot to draw people into malicious websites that steal their info or offer downloads laced with malware. Meta claims it's spotted some of their work.

"As part of our most recent work to protect people and businesses from malicious targeting using ChatGPT as a lure, since March 2023 we've blocked and shared with our industry peers more than 1,000 malicious links from being shared across our technologies and reported a number of browser extensions and mobile apps to our peer companies," Nathaniel Gleicher, head of security policy and Ryan Victory, malware discovery and detection engineer at the social media platform, declared in a statement.

Security engineers at Meta detected various malware strains, such as Ducktail or NodeStealer, hiding in ChatGPT browser plugins and productivity tools, and traced them to scumbags in Vietnam and other locations. The Social Network said it has notified other online platforms hosting the malware as well as law enforcement authorities. 

"This complexity makes the threat of business-targeting malware an ecosystem challenge, requiring a concerted effort to combat these malicious operations across our entire industry," Gleicher and Victory explained. "Like spammers, they latch onto popular topics like political news or the latest tech or business productivity tools to hide their malware and trick people into clicking on or downloading it.

"For example, one of the campaigns we recently disrupted leveraged people's interest in  Open AI's ChatGPT to lure them into installing malware. In response to detection by our security teams and also our industry peers, we've seen bad actors quickly pivot to other themes, including posing as Google Bard, TikTok marketing tools, pirated software and movies, and Windows utilities."

In response to these attacks, Meta said it has launched a wizard-like support tool that walks users through the process of detecting and disposing of malware. Facebook has also added security controls for accounts operated by managers of business pages. Administrators should have better control over access rights, and will have to enable two-factor authentication to verify sensitive actions like accessing a credit line or changing administrative accounts. ®