TechResearchers Use Claude to Hack OpenAI's Internal Systems in Under 72 Hours
Security researchers from Hacktron AI used Anthropic's Claude to exploit a remote code execution vulnerability in OpenAI's community forum, gaining access to employee accounts and GitHub. The attack cost under $3,000 in tokens and took less than 72 hours. OpenAI paid a $6,500 bounty.
TLDR
The incident underscores AI's role in accelerating sophisticated attacks even by ethical researchers, exposes real vulnerabilities in AI company infrastructure, and fuels narratives around competitive dynamics between labs, the dual-use nature of powerful models, and security gaps in the AI ecosystem.
Combined views
744
1 Source, first seen 16h ago
Researchers Use Claude to Hack OpenAI's Internal Systems in Under 72 Hours
Security researchers from Hacktron AI used Anthropic's Claude to exploit a remote code execution vulnerability in OpenAI's community forum, gaining access to employee accounts and GitHub. The attack cost under $3,000 in tokens and took less than 72 hours. OpenAI paid a $6,500 bounty.
TLDR
The incident underscores AI's role in accelerating sophisticated attacks even by ethical researchers, exposes real vulnerabilities in AI company infrastructure, and fuels narratives around competitive dynamics between labs, the dual-use nature of powerful models, and security gaps in the AI ecosystem.