OpenAI has shut down its preparedness team, which was responsible for assessing whether the company's models posed serious risks and developing mitigation strategies. The team, which ended operations at the end of July, had investigated scenarios including potential model misuse and autonomous system vulnerabilities. The dissolution comes as the company faces regulatory scrutiny and internal debates about AI safety priorities.

What This Means for Your Business

This signals a potential shift in OpenAI's risk governance at a critical moment when regulators globally are tightening oversight. For enterprises deploying OpenAI's models, this raises questions about how the company will validate safety claims going forward—and whether internal guardrails around model deployment will weaken. Verify your own testing protocols for model behavior before broader rollouts.