This looks like more OpenAI damage control. Since the Pentagon accuses Dario of having a "God-like complex" one can only surmise that the potential future guardrails Altman claims they can employ are purely optional for the Pentagon when push comes to shove.

But this made me chuckle:

>U.S. intelligence agencies including the C.I.A., which uses Anthropic’s A.I. technology,

...just cause it has amusing parallels with those stories about how the Pentagon were supporting the Kurds and the C.I.A was propping up ISIS in Syria. Now we have their AI agents fighting out in the field to look forward to in the future!

With Anthropic it's just the rest of the world that is fucked, with OpenAI it's the whole world including the US that is fucked. How relieving.

I simply don't believe that "the only thing that remained was agreeing on a few words", regardless of what sources may have told the NYT. That's not compatible with any of the Trump regime's bombastic public statements. Either they were never negotiating in good faith at all, or Hegseth intended to unilaterally declare alterations to the deal once Anthropic announced it was resolved.