“Anthropic plans to meet with global policymakers and rival labs in the coming months to coordinate AI safety thresholds.”