Numerous tech companies are vying to harness the power of AI for a new generation of web browsers. Probably the most prominent is Perplexity's Comet, which it describes as a "personal assistant and thinking partner" while you surf the web.

Unsurprisingly, that approach can have enormous cybersecurity implications. As privacy-focused browser company Brave noted in a blog post last week, it's alarmingly easy for bad actors to trick a browser AI into following malicious instructions embedded in publicly available content.

The vulnerability, known as an indirect prompt injection attack, is terrifyingly simple.

"The vulnerability we’re discussing in this post lies in how Comet processes webpage content," the blog reads. "When users ask it to 'Summarize this webpage,' Comet feeds a part of th

See Full Page