Anthropic Eases A.I. Safety Rules Amid Competition and Pentagon Pressure

Anthropic announced it is loosening its core AI safety commitments, replacing its binding Responsible Scaling Policy with a more flexible, nonbinding one.