Security & Privacy in Agentic Browsing
Summary
A series of blog posts detailing vulnerabilities in agentic AI browsers, specifically focusing on indirect prompt injection attacks via hidden HTML and screenshots.
Key quotes
The attack we developed shows that traditional Web security assumptions don't hold for agentic AI, and that we need new security and privacy architectures for agentic browsing.
This page serves as an index for a blog series regarding the security implications of AI-driven browsing. It links to research on prompt injection vulnerabilities in Opera Neon and Perplexity Comet.