The people warning us about AI are also building it
Anthropic, an AI company that has pushed for stronger safety protocols, faced a White House clampdown on its latest models over security risks, highlighting the tension between AI safety advocacy and …