Security researchers at Hacktron AI used Anthropic’s Claude to chain together vulnerabilities in OpenAI’s systems, gaining access to employee ChatGPT accounts and internal repositories. The team exploited an image-upload bug in the Discourse forum software along with a flaw in OpenAI’s employee-account validation, completing the operation within 72 hours in late July. OpenAI paid the researchers $6,500 through its bug-bounty program and has since patched the vulnerabilities.
