Nvidia's CEO just said what nobody in AI wanted to say out loud. "If the product is not safe, just don't release it. It's that simple." Jensen Huang says if Anthropic and OpenAI keep telling the world their models are dangerous, they obviously know it, so they should stop