A three-month in-person fellowship at the Center for AI Safety in San Francisco for researchers examining the societal dimensions of advanced artificial intelligence from economics, law, international relations and adjacent disciplines such as political philosophy and security studies. Eligibility is set at professors, PhD or JD holders, and researchers with equivalent experience. Fellows receive $25,000 plus covered travel and daily meals, and are expected to produce substantive interim output — blog posts, draft papers, theoretical models or empirical results — by the programme's end. The 2026 cohort ran 1 June to 21 August with a 24 March deadline.