Montréal’s AI safety coworking and event space.
Desks for people working on reducing risks from AI, and the room where AI Safety Montréal holds its talks, reading groups and hackathons.
3813 rue Saint-Denis, on the Plateau. From CA$10 a day.
What you get
A seat
Ten seats, some with monitors, in a high-ceilinged room, plus a kitchenette and a balcony.
Introductions
Tell us what you work on, and we’ll introduce you to people here and elsewhere who work on it.
Membership
Students, people between jobs and researchers without funding pay the Community rate of CA$10 a day. Cancel any time. How membership works
Who works here
Two members, Robert Graham and Yariv Barsheshat, co-authored Innocuous-Seeming Data, Latent Ideology (arXiv, July 2026). In June, Ω Labs hosted the AI Control Hackathon, organized with HackOS and sponsored by Redwood Research, NVIDIA, Rootly and Rubric Labs: about 50 people over a week, on site and online.
The space
Events coming up
Nearly every event on the AI Safety Montréal calendar happens in this room, and most are open to everyone.
- Safe Agentic Workflows: From First Guardrails to Adversarial Oversight
- Lunch talk: Alan Chan presents What If Automating AI R&D Triggers an Intelligence Explosion?
- AI Safety Technical Reading Group – Value Leakage: An LLM's Answers Are Silently Shaped by Its Own Values
- L'IA nous ment-elle?
- AI Collusion Research Sprint
- This Month in AI Safety / Les nouvelles du mois en sécurité de l'IA