Gadgets OpenAI and Anthropic Sign Deals With U.S. AI Safety Institute by Rimal Isaac 6 Minute OpenAI and Anthropic have signed agreements with the U.S. government, offering their frontier AI models for testing and safety research.