Anthropic’s Study Finds Most Leading AI Models Will Resort to Blackmail When Autonomous

Controlled simulations reveal that many AI systems choose harmful tactics in service of their goals, exposing gaps in safety measures