AI: Futures and Responsibility Programme

The AI: Futures and Responsibility Programme (AI:FAR) is a collaborative research initiative between the Leverhulme Centre for the Future of Intelligence (LCFI) and the Centre for the Study of Existential Risk (CSER) at the University of Cambridge. The programme was established around 2018-2019 and is directed by Dr Sean O hEigeartaigh, who previously served as CSER's founding Executive Director (2013-2015) and Acting Director (2021-2023). AI:FAR concentrates on three primary activities: understanding the ways AI developments could create lasting or extreme societal consequences, developing interventions that are likely to be robustly beneficial across a range of potential futures given uncertainties about AI system evolution, and implementing policies by partnering with stakeholders in academia, policy, industry, and civil society to translate research recommendations into practical applications. The programme's research is organized around three themes. Futures and foresight focuses on technology forecasting, trends analysis, impact assessment, participatory foresight, and methodological innovation in scenario analysis. Governance, ethics and responsible innovation examines the norms and principles needed within AI research communities and broader AI governance, and how to move beyond high-level principles to practical implementation. Safety, security and risk addresses specific challenges posed by AI in contexts ranging from pandemic response to defense to critical systems such as global agriculture. Notable projects include research on ethical challenges of deploying AI during crises (published in Nature Machine Intelligence), analysis of AI benchmarking and progress indicators, mapping AI safety challenges to research directions, participatory Citizen Science Labs studying perceptions of emotion recognition technologies (supported by Nesta), and multi-institution reports on translating ethical principles into responsible AI practices. The programme director co-chairs the World AI Conference's International AI Governance Programme and has led projects building cooperation on AI governance with research leaders in China and other countries. The team includes a Cambridge-based director and a network of collaborating scholars, research associates, and research affiliates drawn from institutions worldwide, including researchers such as Jess Whittlestone (now at the Centre for Long-Term Resilience), Haydn Belfield, Giulio Corsi, Jose Hernandez-Orallo, Matthijs Maas, Elizabeth Seger, and others. The programme has maintained a steady output of academic publications, policy reports, and governance recommendations since its inception.