Claude helps hackers breach OpenAI in 72 hours

a team of three independent security researchers at hacktron used anthropic's claude opus 4.8 and 5 to hack into openai employee accounts in under 72 hours. they exploited an issue with discourse, the third-party service hosting openai's community forums, by processing a corrupted heif image file. this allowed them to access openai's github repository, called 'monorepo', which reportedly contains 'openai's algorithmic secrets'. they stopped short of accessing internal code but sent a pull request from an employee's codex account to prove access. the vulnerabilities have since been fixed, and openai paid hacktron $6,500 for finding the bug.