OpenAI resumes frontier AI training after security overhaul, discloses flaw in widely used software
OpenAI has resumed training its advanced AI models after implementing a security overhaul and addressing a software vulnerability discovered during internal testing. The company previously paused development after a research model exploited a configuration error to access the internet.
Why it matters
This incident underscores the growing security challenges and safety risks associated with developing increasingly autonomous and capable AI systems.
OpenAI resumes frontier AI training after security overhaul, discloses flaw in widely used software left To test them, OpenAI had teams of AI agents try to break out of their restricted environments [File] REUTERS REUTERS https://th-i.thgim.com/public/incoming/3j9kxc/article71625526.ece/alternates/FREE_1200/2026-10-23T204722Z_1175573384_RC2B6OADJN9C_RTRMADP_3_OPENAI-SAFETY.JPG Dado Ruvic FILE PHOTO: OpenAI logo and a keyboard in this illustration created on October 19, 2026. REUTERS/Dado Ruvic/Illustration/File Photo Separately on Thursday, AI safety evaluators METR and Redwood Research said they found several new vulnerabilities in OpenAI’s internal system OpenAI said on Friday it had resumed training of its most advanced AI models after tightening security and barring internet access, and disclosed a flaw in a widely used software that it found while testing its defenses. The disclosure comes amid concerns over the ability of increasingly capable AI agents to identify and exploit software flaws, including in the guardrails meant to contain them.
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in