If Anthropic and OpenAI are serious about a global government supported AI slowdown then they should reveal where their models are internally because if that plan goes ahead they have to do that anyways it would build trust, show that they are committed and not just larping