scalable oversight

Jan Leike joins Anthropic to lead the new ‘superalignment’ team, emphasizing AI safety and security. The team focuses on scalable oversight and automated alignment research to ensure robust AI development aligned with human values.