Trump administration asks to hold OpenAI's next model (www.axios.com)

The Trump administration has requested that OpenAI restrict the release of its upcoming model, GPT-5.6, to a select group of government-approved partners prior to a broader launch, citing security concerns. This unprecedented move signifies the U.S. government's proactive stance on AI safety, emphasizing the need for a structured framework to evaluate and test the capabilities of advanced models. OpenAI CEO Sam Altman acknowledged this limitation in a memo, expressing the company's commitment to collaborating with the government for a more sustainable release approach in the future. The request comes as AI models grow more powerful, with GPT-5.6 exhibiting capabilities comparable to other cutting-edge models like Anthropic's Mythos 5. The administration's intervention highlights concerns regarding the potential misuse of such technology by malicious actors, from cybercriminals to state-sponsored threats. With heightened scrutiny on emerging AI capabilities, the industry faces pressure not only to innovate swiftly but also to ensure robust safeguards are in place. Altman indicated that OpenAI aims to release GPT-5.6 shortly after this limited rollout, reflecting the ongoing tension between technological advancement and regulatory oversight in the AI landscape.