We're pleased to present another of our Poster Highlights at TAIS 2025: "Why Accelerating Human–Machine Interfaces (HMI) Matters: A Synergy–Overhead Framework for AI Oversight"! Roman Yamashina, founder of Cyber Ascension, is presenting a paper highlighting the importance of

TAIS 2025 is only one week away! Come connect with leading AI safety researchers from Japan and abroad. Shape conversations that will influence how we build safer AI systems for our shared future. Join us in Tokyo on Saturday, April 12th - register on Luma:

Upcoming Poster Highlight at TAIS 2025: "Active Teacher Selection for Reinforcement Learning from Human Feedback"! Rachel Freedman is an AI safety researcher, Cooperative AI Foundation PhD Fellow, and ML PhD student at the Center for Human-Compatible AI at UC Berkeley. Key

We're pleased to present another of our Poster Highlights at TAIS 2025: "Desire Homuncularism"! Steve Petersen is an academic philosopher working to widen the bottleneck between technical AI alignment and analytic philosophy. Systems with "real caring" would both be potentially

Our first speaker at TAIS 2025: Ryota Kanai As technology advances at an unprecedented pace, the concept of the global workspace offers a powerful framework for understanding and shaping the future of AI, human cognition, and society. This talk explores how this theory,

Our Final Speaker at TAIS 2025: Adam Gleave Starting by examining risks advanced AI poses from misuse, loss of control and societal destabilization, we'll explore through interactive demos the capabilities of contemporary AI systems to cause harm, and forecast the performance of

TAIS 2025 is only one day away!  Come connect with leading AI safety researchers from Japan and abroad. Shape conversations that will influence how we build safer AI systems for our shared future. Join us in Tokyo on Saturday, April 12th - Doors open at 11:30:

Thank you again for joining us at TAIS 2025! We hope to see you all again next year! #AI #TAIS2025 #AIConference #AISafety

The TAIS 2025 Best Poster was awarded to Antonio-Gabriel Chaćon Menke and Phan Xuan Tan for their work "How Effective Is Constitutional AI in Small LLMs? A Study on DeepSeek-R1 and Its Peers". Their paper explores CAI's self-critique mechanism on small, uncensored 7-9B parameter

TAIS 2025 may have come and gone, but our memories of it haven't! Once more we'd like to thank our sponsors: Noeon Research , Ashgro, and AI Alignment Network (ALIGN) , as well as our speakers: Ryota Kanai Ryota Kanai of ARAYA.org / ALIGN, and Adam Gleave Adam Gleave of

TAIS 2026 is officially returning in May—this time in the UK! Want to submit your research? Starting today, October 1st, submissions open here: tais2026.cc/paper-submissi… Stay tuned for details, dates, and more news to follow soon.

The date for TAIS 2026 has been set! Join us Thursday, 14th May 2026 at Oxford Martin School (UK), in partnership with the Oxford Martin AI Governance Initiative. Paper submissions are open until January 31, 2026: tais2026.cc/paper-submissi… This one-day academic conference will feature technical talks and

We're delighted to announce our first speaker for TAIS 2026—Oliver Sourbut Oliver Sourbut of the Future of Life Foundation! Join us in Oxford next Spring, 14th May 2026!

Proudly announcing our second speaker for TAIS 2026—Seán Ó hÉigeartaigh Seán Ó hÉigeartaigh of Future Intelligence! Join us in Oxford next Spring, 14th May 2026!

If you would like to showcase your AI Safety research at TAIS 2026, paper submissions are due by 31 January 2026. Notification of acceptance will be 15 February 2026. Submit your paper here: tais2026.cc/paper-submissi…

Starting off the New Year with our third speaker announcement for TAIS 2026 — Markus Anderljung Markus Anderljung of GovAI!

We hope everyone had a lovely holiday season and New Year! Final reminder that paper submissions are due by 31 January 2026.  If you'd like to showcase your AI Safety research at TAIS 2026, please submit here: tais2026.cc/paper-submissi…

Thrilled to announce our fourth speaker for TAIS 2026 — Sara Bernardini Sara Bernardini, Professor of Artificial Intelligence at the University of Oxford! Join us in Oxford, 14th May 2026!

Struggling to make that ICML submission on time? By popular demand, we've extended the deadline for paper submissions to February 15. Submit your paper here: tais2026.cc/paper-submissi…

Excited to announce our fifth speaker for TAIS 2026 — Victoria Krakovna Victoria Krakovna, research scientist on the Alignment team at Google DeepMind and a co-founder of the Future of Life Institute Future of Life Institute! Join us in Oxford, 14th May 2026!