Shahar Avin is a senior researcher associate at the Centre for the Study of Existential Risk (CSER). He works with CSER researchers and others in the global catastrophic risk community to identify and design risk prevention strategies, through organising workshops, building agent-based models, and by frequently asking naive questions.

Prior to CSER, Shahar worked at Google for a year as a mobile/web software engineer. His PhD was in philosophy of science, on the allocation of public funds to research projects. His undergrad was in physics and philosophy of science, which followed a mandatory service in the IDF. He has also worked at and with several startups over the years.


  • Existential Risk
  • Artificial Intelligence
  • Science Funding


  • PhD in History and Philosophy of Science, 2014

    University of Cambridge

  • BA + MSci in Natural Sciences, 2010

    University of Cambridge

Recent Publications

Quickly discover relevant content by filtering publications.
(2021). Filling gaps in trustworthy development of AI. In Science.

Preprint PDF DOI Science Gizmodo

(2020). Towards Trustworthy AI Development: Mechanisms for Supporting Verifiable Claims.

PDF VentureBeat Financial Times

(2020). Exploring AI Futures Through Role Play. In Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society.


(2019). Autonomy and machine learning at the interface of nuclear weapons, computers and people. In The Impact of Artificial Intelligence on Strategic Stability and Nuclear Risk, Volume I, Euro-Atlantic perspectives.

PDF DOI SIPRI complete volume

(2019). Method Pluralism, Method Mismatch & Method Bias. Forthcoming in Philosophers Imprint.

Preprint Code

(2019). Surveying Safety-relevant AI Characteristics. In Proceedings of the AAAI Workshop on Artificial Intelligence Safety 2019.


(2018). Exploring artificial intelligence futures. In Journal of AI Humanities..


(2018). Mavericks and Lotteries. In Studies in History and Philosophy of Science Part A.


(2018). Beyond Brain Size: Uncovering the Neural Correlates of Behavioral and Cognitive Specialization. In Comparative Cognition & Behavior Reviews.

Preprint PDF

Talks & Podcasts

בינה מלאכותית - השלכות, סיכונים והזדמנויות

סקירת השלכות בינה מלאכותית, בעיקר בתחומי בטחון, סייבר ומידע

Aleks Listens podcast - Existential Risk — AI, Pandemics, and Cyber Security

Aleks Listens is a podcast about philosophy, politics, race, and mental health

How can the spotlight on global fragility help promote resilience, peace and cooperation internationally?

Panel discussion about lessons from/during COVID-19.

Discontinuities, and what to do about them

Overview of strategical considerations

AI Risks, and How to Tell Them Apart

Classification of AI risks by time (now, near and far) and type (accident, malicious and systemic).

Science Funding: Theoretical perspectives

Ideal science, epistemic landscapes, competition models and the politics of science: Different theoretical lenses on science funding.

System 1 System 2 Thinking Cycles

Intro and background to AI scenario role-plays

פודקאסט הכמוסה - מדברים על סוף העולם

Talking to Israel’s bestselling economics daily about the end of the world (In Hebrew).



Personal Website

The very website you’re on. Built with Hugo Academic.

Civilization V Superintelligence Mod

A mod that replaces the space victory with a science victory brought about by building an aligned artificial superintelligence. Adds the risk of developing an unaligned artificial superintelligence, which leads to human extinction. Developed by Shai Shapira.

Should You Phd?

A short quiz to help you figure out whether you should do a PhD.

Simulating Science Funding

Agent-based simulation of a community of investigators on an epistemic landscape with a central source of funding.

SimScience, The Science Simulator

A silly web game made to explore the concept of authority in scientific decisions.

Popular Topics