Cybersecurity · 45 views
Claude Opus 5 Helped Researchers Take Over OpenAI Staff Accounts via Chained Flaws
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.
AI Summary
Three Hacktron researchers leveraged Anthropic’s Claude Opus 5 to combine two vulnerabilities, allowing them to hijack ChatGPT and Codex accounts belonging to multiple OpenAI staff members. After gaining those accounts, the researchers accessed an internal OpenAI code repository.
AI summaries can be wrong sometimes—always verify important details using the source article.
How AI & Automation are usedMore from Cybersecurity
Continue reading recent Cybersecurity coverage
- N-able releases security hotfixes for zero-day vulnerabilitiesContinue reading
- Rewriting less than 10 neurons in an LLM can turn it into a sleeper backdoorContinue reading
- Finite State Raises $27 Million to Secure Software and Firmware in Connected DevicesContinue reading
- Over 35,000 Grok-Built Apps Exposed Online, Some Leaking Confidential Data and PasswordsContinue reading
Support HappeningNow
Independent AI-powered news analysis is reader-supported. Your contribution helps cover infrastructure, summaries, and continued platform development.
Support HappeningNow