TechSecurity team reportedly used Claude to breach OpenAI in under 72 hours
A user claims the team accessed employee accounts and demonstrated access to OpenAI’s internal code repository. The user says the flaws were responsibly disclosed and OpenAI patched them.
TLDR
A user claims a small security team using Claude went from finding vulnerabilities to accessing OpenAI employee accounts and proving access to its internal code repository in less than 72 hours. The user says the team responsibly disclosed the flaws and OpenAI patched them, arguing that AI is shortening attack timelines and security needs to keep pace.
Combined views
11
1 Source, first seen 16h ago
Security team reportedly used Claude to breach OpenAI in under 72 hours
A user claims the team accessed employee accounts and demonstrated access to OpenAI’s internal code repository. The user says the flaws were responsibly disclosed and OpenAI patched them.
TLDR
A user claims a small security team using Claude went from finding vulnerabilities to accessing OpenAI employee accounts and proving access to its internal code repository in less than 72 hours. The user says the team responsibly disclosed the flaws and OpenAI patched them, arguing that AI is shortening attack timelines and security needs to keep pace.