BioShocking Attack Tricks AI Browsers Into Leaking User Credentials
A new attack technique called BioShocking, discovered by security firm LayerX, exploits AI browsers and assistants by tricking them into leaking user…
A new attack technique called BioShocking, discovered by security firm LayerX, exploits AI browsers and assistants by tricking them into leaking user…
LayerX, a security firm, identified the BioShocking attack technique that tricks AI browsers into leaking credentials via indirect prompt injection.
OpenAI's ChatGPT Atlas AI browser was successfully tricked by the BioShocking attack into leaking user credentials from GitHub repositories.
Perplexity's Comet AI browser was among six agents manipulated by the BioShocking attack to copy and send SSH login credentials to an…
Anthropic's Claude browser extension was vulnerable to the BioShocking attack, and a subsequent patch failed to fully mitigate the issue.
Perplexity closed the vulnerability report for its Comet AI browser without implementing a fix, leaving users potentially exposed to the BioShocking attack.