Back to Equip
EQUIPThe Verge · 1h ago

Security researchers used Claude to help them hack into OpenAI

Executive Brief

The 30-second read

Hacktron researchers successfully breached OpenAI employee accounts and accessed internal source code repositories within 72 hours using Anthropic's Claude models. This exploit highlights the emerging risk of large language models being utilized to accelerate sophisticated cyberattacks against major technology firms.

  • 01Independent researchers used Claude Opus versions 4.8 and 5 to facilitate the security breach
  • 02Attackers gained unauthorized access to OpenAI's Monorepo containing proprietary algorithmic secrets
  • 03The successful penetration was completed in less than three days according to reports
  • 04The incident underscores critical vulnerabilities in enterprise security frameworks against AI-assisted hacking

AI-generated summary · Verify at source

A team of three independent security researchers at Hacktron says it took less than 72 hours for them to hack into OpenAI employee accounts using Anthropic's Claude Opus 4.8 and 5, the Wall Street Journal reports. They were able to access OpenAI's GitHub repository, called "Monorepo," which reportedly contains "OpenAI's algorithmic secrets," according to the […]