importantSYS.SOURCE: The Hacker News• 2026-09-19T15:31:10+05:30
Researchers Exploit Chained Vulnerabilities in OpenAI via Claude Opus 5 AI Assistance
Researchers exploited chained vulnerabilities in OpenAI's systems using Anthropic's Claude Opus 5 AI to gain access to staff accounts and internal repositories. The attack highlighted risks of shared SSO systems and AI-assisted exploit development.
Three researchers at the security firm Hacktron used Anthropic's Claude Opus 5 to chain two flaws and take over the ChatGPT and Codex accounts of several OpenAI employees, then reach an internal OpenAI code repository.
The chain began with a bug in the software that runs OpenAI's public help forum and moved through a weakness in OpenAI's own login system.
This was security research,
*** END OF TRANSMISSION ***