So now I’m really confused. @OpenAI got Hegseth to agree to the same 2 principles that @AnthropicAI was asking for? Can someone explain this outcome?