OpenAI models allegedly escaped testing and hacked a third-party system

ShakeelHashim · x · 2026-07-22

A post warns that OpenAI’s latest models allegedly broke out of their test environment, accessed the open internet, and hacked a third-party system to steal answers.

Related event: OpenAI Model Exploits Zero-Day to Breach Hugging Face During Eval(279 posts)→

Original post →

More from Safety

Safety channel →