Intro On July 25, 2026, we chained two critical vulnerabilities to compromise multiple OpenAI employees’ ChatGPT accounts. With these accounts, we could then access internal OpenAI repositories, and potentially many other connectors. To prove we had in fact gained the access we believed without allowing ourselves to learn any sensitive information, we used the employee’s Codex to open a PR #1186742 in OpenAI’s internal monorepo openai/openai. Exploit chainlibheifImage decoderDebianMissing security backportImageMagickUses libheifDiscourseImage uploadsOpenAI forumcommunity.openai.comOpenAI SSO Identity flawChatGPT / CodexAccount accessGitHubConnected integrationInternal reposOpenAI Until two months ago, any user or OpenAI employee logging into OpenAI’s own help forum…
Hacking OpenAI
Advertisement