Anthropic is defaulting Claude to an 'auto mode' that acts autonomously, saying it judges risk better than humans do.
Continue to AI University →