A Call for Open Science in AI Safety
We call on frontier model developers to share their AI safety methods and research with the global community, to enable independent scientific scrutiny and advancement.
The safety of increasingly powerful AI systems should be documented with detailed descriptions of methods and evaluations that independent researchers can examine, challenge, reproduce and improve.
Transparency will make a big difference for both assessing and improving the safety of AI models.
Assessing AI safety will be facilitated by sharing sufficient detail about the safety-relevant aspects of each frontier model—including evaluations, safety-training recipes, relevant code and data, and evidence of desired and undesired behavior. This will help advance the scientific understanding of model behavior and the effectiveness and limitations of safety measures.
Improving AI safety will be facilitated as transparency will allow research efforts and funds—across academia and industry—to be spent towards the most promising directions in a targeted way. Drawing on the expertise and active contribution of the global scientific community, sharing safety best-practices helps improve the safety of all AI models, whether closed or open, irrespective of geographic origin. This can accelerate progress in AI safety, allowing safety research to keep pace with the fast advances in AI capabilities.
In cases where releasing particular information could itself create a credible security or misuse risk, exceptions can be made. Such exceptions should be specific and proportionate, rather than becoming a general reason for keeping safety evidence closed.
Signatories
Initiated by:
Martin Jaggi (EPFL), Robert West (EPFL), Philip Torr (Oxford), and Anna Hedström (ETH Zurich).
Published: 18th Sept 2026
Will be updated from time to time.
All endorsements are in a personal capacity.
Affiliations are provided for identification only.
- Martin Jaggi, EPFL
- Robert West, EPFL
- Philip Torr, Oxford
- Anna Hedström, ETH Zurich
- Samuel Simko, ETH Zurich
- Hanna Yukhymenko, ETH Zurich
- Maria Ios Glarou, EPFL
- Imanol Schlag, ETH Zurich
- Jonas Hertner, Independent lawyer
- Mallory Wittwer, EPFL
- Gabriele Sarti, Northeastern University & Parallax
- Roland Aydin, Saarland University
- Sébastien Marcel, Idiap research institute
- Chengkun Li, EPFL
- Elie Bussod , EPFL
- Grégoire Clément
- El Mahdi Chayti, EPFL
- Afshin Khadangi, University of Luxembourg
- Andrei Semenov, EPFL
- Namhoon Lee, POSTECH
- Marcel Blattner, HSLU
- Marc-Antoine Allard, EPFL - ILLUIN Tech
- Deblina Bhattacharjee, University of Bath, UK
- Vivek Sharma, MIT, Harvard Medical School/MGH
- Adrien O’Hana, Surelio.ai
- Daniel Nissani, Mozilla.ai
- Andrei Kuchravy, HES-SO Valais Wallis; Surelio SA
- Erwin Fang, Hochschule für Wirtschaft Zürich
- Douaa Kdidar, EPFL
- Xin Chen, Cynthia, ETH Zurich
- Valentin Conrad, EPFL
- Utkarsh Gupta, Delhi Technological University
- Jianing Zhu, UT Austin
- Aman Sinha, Université de Genève
- Manoel Horta Ribeiro, Princeton University
- Saibo Geng, EPFL
- Lie He, SUFE
- Addison Kristanto Julistiono, EPFL
- Tommaso Cerruti, ETH Zurich
- Xinxian Ma, EPFL
- Akhil Arora, Aarhus University
- Andy Arditi, Northeastern University
- Dominik Winterer, University of Manchester
- François PORTET, Université Grenoble Alpes
- Mary-Anne (Annie) Hartley, EPFL
- Lennig Pedron, Trust Valley / EPFL Innovation Park
- Amir Zade, EPFL
- Dun Li Chan, Independent
- Lukas Fluri, ETH Zurich
- Maria Brbic, EPFL
- Lennart Finke, ETH Zürich
- Abhisek Dash, Max Planck Institute for Software Systems
- Alain Milliet, Richemont International SA
- Margarita Sagitova, EPFL
- Sushovan Jena, IIT Mandi
- Ken Holstein, EPFL
- Darpan Aswal, Université Paris-Saclay
- Mirjana Stojilović, EPFL
- Bartol Bućan, EPFL
- Seng Zhen Hong, NUS
- Arnau Padrés Masdemont, Qualcomm
- Imran Zualkernan, American University of Sharjah
- Tianhao Dai, EPFL
- Gerard de Melo, University of Potsdam
- Matea Tashkovska, EPFL
- Fay Elhassan, EPFL
- Shaokai Yang, University of Alberta
- Clément Dumas, Astra fellowship
- Grégory Mermoud, HES-SO Valais
- Michael Schumacher, University of Applied Sciences and Arts Western Switzerland (HES-SO)
- Julian Minder, EPFL
- Maksym Andriushchenko, ELLIS Institute Tübingen
- Anish Acharya, AWS
- Rudolf Debelak, EPFL
- Sabine Wildemann, aiLights Association
- Ejiro Onose, Ethos
- Yassine El Aouad, EPFL
- Saibal Dutta, Independent Researcher
- Sivaram A, Phygitalytics
- Eva Thelisson, AI Transparency Institute
- Ying Zhang, Copenhagen Business School
- Bravish Ghosh, IIT Madras
- Gustavo Sandoval, New York University
- Dirk Wulff, MPI for Human Developmet
- Simin Fan, EPFL
- Bashir Ansari, Indian Institute of Technology Dhanbad
- Shunchang Liu, EPFL
- Giovanni Colavizza, University of Copenhagen
- Fazl Barez, Oxford
- Ben Chaabane Elyes, EPFL