Overview
As one of the world’s leading AI nations, with a long track record and deep expertise in AI, Canada has a critical role to play in advancing the world’s understanding of the risks associated with advanced AI systems and how to mitigate them.
Building on decades of investment in research leadership and our strong commitment to responsible AI, the CAISI Research Program at CIFAR advances Canadian-led fundamental and applied research on the technical and social-impact challenges of AI Safety.
The CAISI Research Program at CIFAR will:
- Take a multidisciplinary approach to understanding AI safety and how to mitigate the risks.
- Address both short- and long-term risks of frontier AI systems.
- Provide opportunities for both rapidly deployed projects and longer-term collaborations to develop new tools and approaches.
- Advance both fundamental and applied research.
- Build capacity and a community of researchers working on this effort across Canada.
Safety research news
AI and Society
Government of Canada announces Canadian AI Safety Institute
November 12, 2024
AI and Society
Beyond privacy: it’s time for a rights-based approach to regulating AI for children
April 24, 2024
AI and Society
Policy advice for navigating regulation in the age of AI: think bigger
November 21, 2023
AI Safety
Foutse Khomh, Mila: Making the world’s AI safer
October 09, 2023
AI Safety
Sheila McIlraith, Vector Institute: Building safe AI for the generations to come
October 09, 2023
Advances
Believe the Impossible: The Future of… Fairness in AI
September 12, 2022
AI and Society
A Culture of Ethical AI: Workshop report
August 03, 2022
AI and Society
Recommendation algorithms could contribute to inequality
June 03, 2021 Get involved
Apply to be a member of the CAISI Research Council
The CAISI Research Council is responsible for providing advice and guidance to CIFAR in the delivery of the CAISI Research Program. CIFAR is now recruiting members, with three open positions for Members at Large. The Council meets monthly via Zoom and a small annual honorarium is provided. Any active member of the Canadian AI ecosystem from any discipline or sector with interest and expertise in AI safety is welcome to apply by submitting a short statement of interest (max. 250 words) and CV to ai@cifar.ca by 5:00 pm ET, December 16, 2024. Only selected applicants will be contacted.
Stay informed
For ongoing updates on the activities of the CAISI Research Program, subscribe to AICan Bulletin, our monthly newsletter outlining the advances of the Pan-Canadian AI Strategy at CIFAR.
CAISI Research Council
The CAISI Research Program at CIFAR is guided by a council of experts, led by two Co-Directors, to be named.
The Council is comprised of research experts (members at large) and representatives from CIFAR and the National Research Council, drawing on extensive expertise at Canada’s three National AI Institutes: Mila in Montréal, Amii in Edmonton, and the Vector Institute in Toronto.
