The Leverhulme Centre for the Future of Intelligence (CFI) is an interdisciplinary research centre within the University of Cambridge that studies artificial intelligence. It is funded by the Leverhulme Trust. As of 2026, the centre is led by executive director Rachel Adams, with Stephen Cave serving as academic director.
The Centre brings together academics from the fields of computer science, philosophy, social science and others. The centre works with the Oxford Martin School at the University of Oxford, Imperial College London, and the University of California, Berkeley and has a memorandum of understanding with the Coral Bell School of Asia Pacific Affairs at the Australian National University.
Research Programmes
CFI research is structured in a series of programmes examining diverse aspects of AI development and governance. The programmes address topics ranging from algorithmic transparency to the implications of AI for democratic institutions. Key initiatives include AI: Futures and Responsibility, AI: Trust and Society, Kinds of Intelligence, AI: Narrative and Justice, and Philosophy and Ethics of AI. These programmes engage with foundational questions in Machine learning and Deep learning, such as the nature of general intelligence and the design of transparent systems, while also exploring broader societal dimensions like public trust and narrative framing.
The centre's collaborative approach extends beyond Cambridge, linking scholars with partners at the Oxford Martin School, Imperial College London, and the University of California, Berkeley. A memorandum of understanding with the Coral Bell School of Asia Pacific Affairs at the Australian National University broadens its international reach, facilitating cross-cultural perspectives on AI ethics and policy.
Animal-AI Olympics
In July 2019, the centre released the Animal-AI Olympics competition, a novel benchmark that employed tests typically used to evaluate animal intelligence. The initiative aimed to assess the capabilities of neural networks and other AI systems against cognitive challenges drawn from comparative psychology, such as object permanence and spatial reasoning. The competition provided insights into the differences between biological and artificial cognition, informing research on Generative AI and system robustness.
Institutional Context
The Leverhulme Centre for the Future of Intelligence operates within a broader landscape of research organisations focused on AI's long-term implications. It maintains connections with institutions such as the Centre for the Study of Existential Risk at Cambridge, the Future of Humanity Institute at Oxford, the Future of Life Institute, and the Machine Intelligence Research Institute. These collaborations foster interdisciplinary dialogue on topics including the development of large language models and the potential societal impacts of advanced AI technologies.
The centre's work intersects with technical advances in fields like Transformer (architecture) architectures and Generative AI, providing philosophical and ethical analysis that complements research at industrial labs such as Google DeepMind and OpenAI. By engaging with both academic and industry stakeholders, the CFI contributes to ongoing debates about responsible AI development and governance.
Leadership and Funding
The centre is supported by the Leverhulme Trust, a major UK funder of academic research. As of 2026, its leadership includes executive director Rachel Adams and academic director Stephen Cave. The team comprises researchers from philosophy, computer science, social sciences, and related disciplines, reflecting the centre's commitment to holistic analysis of AI. This structure enables examination of topics from technical design choices to ethical frameworks for deployment.
Through its programmes and collaborations, the centre addresses pressing questions about how AI systems should be built, evaluated, and integrated into society, serving as a hub for critical scholarship in the field.