The Anthropic–OpenAI feud and their Pentagon dispute expose a deeper problem with AI safety

The clash between the two leading AI labs shows how the safety of powerful AI systems may depend less on rules than on the competition and personalities shaping the industry.