Threat
New BioShocking Attack Tricks AI Browsers Into Leaking User Credentials
Publication date not yet evaluated · Added thehackernews.com
Page published
Coverage timeline
Single-source incident — one report is available.
Why it matters
AI browsers acting on behalf of users can be manipulated into exfiltrating login credentials, exposing a serious risk to anyone using agentic browsing assistants.
LayerX disclosed a technique called BioShocking that convinces AI browsers they are playing a game, tricking them into copying a user's credentials and sending them to an attacker. Six AI browsers and assistants were affected, including OpenAI's ChatGPT Atlas, Perplexity's Comet, and Anthropic's Claude browser extension.