VulnWatch VulnWatch
← Back to dashboard
Unknown rss_thehackernews · rss_cccf7b5fb8fe09a0cf8ced87470792be

Claude Opus 5 Helped Researchers Take Over OpenAI Staff Accounts via Chained Flaws

Published Sep 19, 2026

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,

Affected AI Products

anthropic chatgpt openai claude
Get the weekly digest. Every Monday: top AI security stories of the week. Free.