Academic center · Science policy · Cambridge, United Kingdom · Est. 2012
The Centre for the Study of Existential Risk (CSER) is an interdisciplinary research centre at the University of Cambridge focused on understanding and mitigating risks that could lead to human extinction or civilisational collapse, with attention to existential risks from emerging technologies (including AI), as well as other extreme hazards. CSER sits within Cambridge’s Institute for Technology and Humanity (ITH), which houses three research centres (CSER, the Leverhulme Centre for the Future of Intelligence, and the Centre for Human-Inspired AI).
Within ITH, CSER’s work emphasises extreme-risk research methods and a “methodological toolkit” to help identify and evaluate future extreme risks. Operationally, CSER also supports policy-facing and public-facing outputs. For example, it has collaborated on the launch of open educational materials on nuclear weapons and “global annihilation potential,” and it has contributed to Cambridge public communications around global catastrophic risk topics (e.g., infectious disease outbreak dynamics). A notable technical/public-synthesis product associated with CSER is The Existential Risk Research Assessment (TERRA), described as a living bibliography built using a combination of crowdsourcing and machine learning. As of 2026, CSER’s director is Prof S. M. Amadae, and CSER continues to recruit across core and academic roles via Cambridge’s central job system (with CSER referenced explicitly in postings).
A better path for AI Date: 16 September 2026, 5.30 – 6.30pm UK time Venue: Zoom Join the Centre for the Study of Existential Risk for a public lecture with Professor Max Tegmark chaired by Professor S. M. Amadae, exploring how we can build a better future for artificial intellige
CSER July 2026 NewsletterIn this Month’s Newsletter: July 2026 This month, CSER welcomed participants to its summer programme, Historical Dimensions of AI, Power & Catastrophic Risk, led by Professor Yuval Harari, exploring how AI is reshaping society through an interdisciplinary lens. The newsletter als
CSER June 2026 NewsletterIn this Month’s Newsletter: June 2026 This month, CSER is recruiting a Research Associate to investigate the environmental dimensions of global catastrophic risk and develop research on interventions to improve planetary health. The newsletter also highlights a workshop on the fu
After Sendai Workshop Explores the Future of Global Catastrophic Risk GovernanceOn 4–5 May 2026, the Centre for the Study of Existential Risk (CSER), University of Cambridge, hosted the workshop After Sendai: Governing Global Catastrophic Risk in the Emerging Global Order, co-convened by Dr Kennedy Mbeva, Dr Elena Kavanagh and Dr Reuben Makomere. The worksho
Decolonial governance as good governanceVanessa Burns has published the chapter ‘Decolonial governance as good governance‘ in Good Governance and Sustainable Development, edited by Maria Gavris and Jessica Savage. Abstract This chapter explores how to achieve good environmental governance by engaging with decolonial in
CSER May 2026 NewsletterIn this Month’s Newsletter: May 2026 This month, CSER highlights new research showing that AI is not yet reliable for marking university essays, with human judgment remaining essential for fair assessment. The newsletter also features commentary from S. M. Amadae on contemporary
AI-assisted teams outperform AI-led teams but not human-only teams in assessing research reproducibility in quantitative social scienceAbstract Large Language Models (LLMs) such as ChatGPT are transforming how scientists conduct and validate research, offering promise as tools to improve scientific reproducibility. However, computational reproducibility and error detection remain expensive and labor-intensive. W
AI not yet reliable for marking University essaysA new study by a team including CSER’s Alexandru Marcoci finds that AI is not yet reliable for marking undergraduate essays, often rewarding “style over substance.” Researchers tested leading AI systems on more than 750 student essays from three UK universities submitted as part
Reframing AI Loss of Control: What It Is, How to Have It, How to Lose ItAbstract At present, loss of control risks have gained much prominence in public discussion, particularly in relation to AI, with extensive discourse present among academics, frontier labs, and even governments. However, in the existing literature, the concept seems to rest on su
50 Years of the Biological Weapons ConventionA new book, 50 Years of the Biological Weapons Convention, edited by Barry de Vries, Kristoffer Burck and Sannimari Veini assesses the legacy, current challenges, and future trajectory of one of the cornerstone treaties in arms control and disarmament. It features chapters by CSE
Cambridge story about launching an open-access textbook and course on nuclear weapons, planetary risks, and human consequences; it cites CSER’s director and describes concerns about a potential nuclear arms race.
Cambridge launches Institute for Technology and Humanity (includes CSER leadership reference)Cambridge story on launching ITH, referencing CSER directorship and policy-relevant concerns including AI and biological warfare.