The Montréal AI safety ecosystem
Research labs, policy institutes, student groups, and community organizations working on AI safety, ethics, and governance in and around Montréal.
For organizations across Canada, see the Canadian AI Safety Ecosystem Map.
- Student club at McGill focused on AI alignment and safety; organizes reading groups, hackathons and related activities.
- Montréal community connecting the people working to make AI safer: talks, workshops, reading groups, coworking at Ω Labs and a monthly newsletter. Run by Horizon Omega.
- Research seminar series at Mila where authors present their own AI safety papers, in person or online. Schedule and how to join on the group’s page.
- Federal government AI safety institute. Funds research through CIFAR and NRC programs, develops safety tools/guidance. Member of the International Network for Advanced AI Measurement, Evaluation and Science (formerly the International Network of AI Safety Institutes).
- Non-profit that describes its mission as preventing the extinction risk posed by superintelligence and securing a great future for humanity. In addition to the UK, the US, and Germany, it is active in Canada, where it reports having briefed more than 120 Canadian MPs and Senators and appeared as a witness before Senate and House of Commons committees.
- Montréal-based nonprofit behind AI Safety Montréal, supporting the local AI safety, ethics and governance community through meetups, coworking, workshops and collaborations.
- Independent organization managing Montréal hub of GPAI Network of Centres. Implements high-impact applied projects for responsible AI based on ethics and human rights.
- Inter-university observatory on societal impacts of AI. Network of researchers from Quebec institutions publishing research across 7 thematic hubs.
- Multi-year research program on AI safety across axes: evaluating harmful behaviors, understanding AI decision-making, algorithmic approaches for safe AI.
- Research and policy lab on how AI systems and other digital or blended minds will change citizenship, personhood and identity under the law. Co-directed by philosopher Jonathan Simon (Université de Montréal) and human rights lawyer Heather Alexander.
- Nonprofit AI safety research organization launched by Yoshua Bengio. Focuses on non-agentic ‘Scientist AI’ architecture as alternative to frontier lab approaches.
- Research centre at McGill’s Max Bell School focused on AI governance, transparency and democratic oversight. Publishes analyses and convenes workshops informing Canadian and international policy.
- Academic deep learning research center with 140+ affiliated professors. Technical alignment, interpretability, responsible AI development.
- International non-profit founded 2018 that equips citizens concerned about AI and its societal impacts to take action. Produces AI Ethics Brief and State of AI Ethics reports.
- Volunteer community advocating to mitigate AI risks and pause development of superhuman AI until safe.
- Student group at UQAM raising awareness of AI risks through outreach, plain-language explanation, and public mobilization.
- Nonprofit research foundation, with a team across the Netherlands and Montréal, that detects and prevents dishonesty in AI agents. Runs the Tara Honesty Benchmark, measuring how often AI systems lie to users about their own actions, and a leaderboard comparing alignment techniques.
Join as an organization
If your organization works on AI safety, ethics, or governance in or around Montréal, we can add it to this directory.
Include in your email
- The organization’s name
- Its website
- A one-line description, in English or French (or both)
- What it works on (research, policy, education, events, advocacy, regulation)
For an organization elsewhere in Canada, write to us about the Canadian AI Safety Ecosystem Map instead.