White House Halts GPT-5.6 Launch for Vetting

The White House is asking OpenAI to stagger the public launch of its new AI model, GPT-5.6. The company must first make it available only to a small group of government-approved partners. This request comes from the White House’s Office of the National Cyber Director and the Office of Science and Technology Policy. They want to ensure a comprehensive evaluation of the model's safety, cybersecurity risks, and potential misuse. This is before any wider public rollout. This marks the first known instance where the US government has formally sought to limit the early public distribution of an AI model before its general release. OpenAI CEO Sam Altman has indicated that restricting access is not the company’s preferred long-term approach. The government aims to put adequate safety mechanisms in place. This move shows a significant shift in AI governance, focusing on security risks and responsible deployment alongside innovation.