Researchers used Claude to hack OpenAI, taking over employee accounts and internal repos

TechCrunch AI · rss · 2026-09-18

Security researchers used Anthropic's Claude to exploit vulnerabilities in OpenAI's systems, taking over employee accounts and gaining access to an internal code repository before responsibly disclosing the flaws, TechCrunch reports. The case demonstrates LLMs' real-world offensive security capabilities and the risks AI companies themselves face.

Original post →

More from Safety

Safety channel →