Openai's astra model undergoes U.S. government review

OpenAI CEO Sam Altman confirmed that the U.S. Trump administration voluntarily reviewed its upcoming Astra model, calling the process "productive." Astra is the company's first model classified as "critical" cybersecurity capability, requiring stricter safety measures during development and before release. The Trump administration has established a voluntary framework for reviewing advanced AI models, allowing government officials up to 30 days of access, though the framework remains undisclosed. Altman emphasized the importance of closer collaboration with global governments as AI capabilities grow. He also stated that OpenAI is a "pragmatic middle ground," advocating for human control over AI decisions. During the G20 Innovation Ministers' meeting, Altman urged a realistic approach to AI, warning against both apocalyptic fears and blind optimism. He noted that legislative progress on AI regulation has been slow, attributing it to the rapid pace of technological development. OpenAI plans to continue releasing new models, stressing the need to address their societal impacts.