The drama never ends...

OpenAI struck a deal with the US Department of Defense to deploy its models on classified networks, announced minutes after Anthropic's deadline to comply with government demands expired. The author criticizes the move as opportunistic and sloppy, while attempting to steelman Sam Altman's reasoning that it was meant to deescalate a dangerous situation for the AI industry. Key distinctions are drawn between Anthropic's approach (handing over model weights, relying on policy agreements) and OpenAI's approach (retaining API control, enforcing safety via technical safeguards). Anthropic's CEO Dario Amodei called OpenAI's messaging 'straight up lies' and 'safety theater.' The piece also covers Trump's political attacks on Anthropic, the risk that 'lawful use' clauses could be redefined by executive order, and the broader competitive and regulatory dynamics between the two leading AI labs.