BrowseSafe: Understanding and Preventing Prompt Injection Within AI Browser Agents

kept by eddie dec 03, 25

Perplexity AI introduces BrowseSafe, a defense architecture, benchmark, and detection model for securing browser agents from prompt injection attacks.

Notion AI: Unpatched data exfiltration
Notion AI allowed data exfiltration via indirect prompt injection in March 2025, where edits were saved before user approval, enabling attackers to steal sensitive document contents.
also on #ai-security, #prompt-injection
{{IW4QaZoc2}}
Perplexity open-sources Bumblebee, a read-only scanner that checks developer machines for risky packages, extensions, and AI tool configs during supply-chain incidents.
also on Comet
Perplexity tests Messages integration and usage credits
Perplexity is developing a Mac Messages connector for its Comet browser and testing a usage-credits system for Pro users to buy extra capacity without upgrading.
also on Comet
Designing AI agents to resist prompt injection
OpenAI argues that defending against prompt injection in AI agents requires treating attacks as social engineering and designing systems to constrain impact rather than just filtering inputs.
also on #prompt-injection
Security boundaries in agentic architectures - Vercel
Vercel recommends separating agent harness from generated code execution in isolated sandboxes with secret injection to prevent prompt injection and credential theft in agentic architectures.
also on #prompt-injection