Hacktron group claims to have used Claude AI to access OpenAI internal repository
A group named Hacktron claims to have used Anthropic's Claude Opus 5 to gain unauthorized access to an internal OpenAI repository. While one report specified the alleged breach occurred within 72 hours, the core claim of unauthorized access via Claude AI remains unconfirmed by OpenAI or Anthropic.
What every outlet reports
- A group named Hacktron was involved
- Claude AI was used
- The target was an OpenAI internal repository
- The action involved unauthorized access
All 2 articles
pragmatist1
skeptic1