AI Safety Researchers at Anthropic Work to Prevent Potential Societal Harms from Advanced AI Systems

Spoiler: the nine-person team works for Anthropic.