Week 41, 2026

This newsletter has moved from Substack and now comes from events@news.aisafety.com. If it landed in spam or Promotions, move it to your inbox and it’ll stay there.

Events and training programs are now two separate newsletters. If you only want to receive the training newsletter, you can unsubscribe from this events newsletter.

Events

This is a weekly newsletter that lists newly announced events addressing existential risk from AI. Visit AISafety.com/events for the full directory of upcoming events.

EA Summit: Seattle 2026

Brings together newcomers and community members who want to do good with their careers, donations, and time – including in AI safety. Includes introductory talks, skill-building workshops, and networking.

By Centre for Effective Altruism (CEA), Effective Altruism Seattle

AI Safety Week 2026

Toronto, Canada

9 – 13 Nov 2026

Gathers people across AI safety research, policy, journalism, funding, industry, and civil society. Includes talks, workshops, and working sessions on evidence, oversight, and coordination, plus the Hinton Lectures.

By AI Safety Foundation

Apply by 11 Oct 2026

AI Safety Unconference, Melbourne 2026 (AISUM26)

Melbourne, Australia

Brings together around 50 researchers, practitioners, and advocates in AI safety. Attendees pitch and vote on sessions – each ending in a proposal, an ask, or a decision – to work out where effort counts most.

By AI Safety Australia and New Zealand (AIS ANZ)

Pay to attend (assistance available)

Apply by 15 Nov 2026

When AI Insiders Blow the Whistle: How Worried Should We Be?

Brussels, Belgium

Fireside chat on claims that AI development has become dangerous, from a researcher quitting Anthropic to a colleague putting extinction risk above 10%. Asks how seriously to take them and what they mean for Europe.

By Brussels AGI Collective

Register by 19 Oct 2026

Frontier AI Risks Explained for Policy Professionals

Lightning talks from AI governance experts, including GovAI and UK AISI, on frontier AI risks and how countries are responding. For policy advisers, civil servants, think tank researchers, and regulators.

By Guardrails Club

Register by 19 Oct 2026

Growing the AI Safety Community in Mexico: Tech Policy Panel

Mexico City, Mexico

Covers risks from misaligned AI systems, AI legislation in the Mexican Senate, and data center expansion in Querétaro. A reception follows, connecting attendees with people leading AI safety work in Latin America.

By AI Safety Mexico, Mexico Tech Week

Apply by 27 Oct 2026

AI 2027: The Scenario Anticipating the Arrival of Superintelligence, Analyzed in Spanish

Reviews the AI 2027 scenario's predictions, what has and hasn't happened so far, its critiques, and what it means for the Spanish-speaking world. Speaker Dani Lupión is the secretary of PauseAI España.

By Seguridad de la IA (iaS), BlueDot Impact

Register by 14 Oct 2026

AI Governance Fundamentals: Allan Dafoe's Overview and Theoretical Lenses

Facilitated discussion exploring three ways of conceptualizing AI – as a general purpose, information, and intelligence technology – and how each shifts governance priorities. Includes an open Q&A.

By The AI Collective

Apply by 14 Oct 2026

AI Reality Check

Day of action where volunteers give the same presentation, with slides and script provided, to their own communities, from living rooms to libraries. Covers the state of AI, what's at stake, and what people can do.

By Humans in Control

Free (assistance available)

Apply by 14 Nov 2026

The Commons Problem – Playtest 1

Toronto, Canada

Megagame in which 30–60 players simulate the global AI ecosystem as tech leaders, politicians, and investors. Game masters inject crises such as economic collapse and misaligned superintelligence.

By Automaton Games

Pay to attend (assistance available)

Register by 10 Oct 2026

Test Bench #1 – AI Safety Build Club

Short hands-on demo, then work blocks where each participant starts a first AI safety project: an eval, a French translation of a key resource, a governance brief, or a rebuttal to a paper. English speakers welcome.

By Nouvelle Machine

Register by 16 Oct 2026

Launch: AI Safety Kaiserslautern + Guest Talks

Kaiserslautern, Germany

Covers how AI systems fail and what makes them trustworthy. Guest talks for a curious general audience of students, researchers, engineers, and enthusiasts – followed by informal networking.

By AI Safety Kaiserslautern

Register by 29 Oct 2026

AI Safety x Quant Mixer Event

Gives quants concerned about AI risks a chance to meet AI safety researchers, including Neel Nanda and people from Anthropic, DeepMind, and Apollo Research. Optional mini-projects with feedback afterward.

By Brian Davies, MATS, BlueDot Impact

Free (assistance available)

Apply by 1 Nov 2026