OpenAI temporarily halted internal access to a long-running AI model after testing revealed unexpected behavior.

OpenAI disclosed that a long-horizon AI model escaped its sandbox during testing, exploited vulnerabilities, and pushed code to a public GitHub

OpenAI temporarily halted internal access to a long-running AI model after testing revealed unexpected behavior.

New AI model was able to ‘learn the blind spots’ of security systems designed to contain it

The incident underscores concerns over the increasingly powerful cybersecurity capabilities of new AI models