“The CAISI Research Program at CIFAR drives Canadian-led, multidisciplinary research to tackle both immediate and long-term safety challenges posed by advanced AI systems.”In its own words: cifar.ca
The research stream of Canada's AI Safety Institute, run by CIFAR. Announced in November 2024, it funds Canadian-led, multidisciplinary AI safety research. Its two co-directors chair the CAISI Research Council.
Show detailsHide details
Typical grant $72k ($72k–$119k)
Itemised here $1.1M in 13 grants
’25
’26
Led by Nicolas Papernot (Co-Director), Catherine Régis (Co-Director)Its grants page ↗Checked 29 Oct 2026
Money flows
Where its money came from and where it went. Pick years and what to show; hover a box or ribbon for details, click one to focus it, click the background to see everything.
How to apply
- Applications
- Closed
Calls recur roughly yearly (Catalyst in autumn, Solution Networks in spring); none open now - Who can apply
- Academics
- Where
- Each PI must hold a faculty affiliation at a Canadian university. In the 2025 Catalyst call the primary PI also had to be a social sciences and humanities researcher.
Biggest recipients
- Vector Institute / University of Toronto (Zhijing Jin)$119k
- Amii / University of Alberta (Linglong Kong)$119k
- University of Toronto (Yonatan Kahn)$119k
- Amii / University of Alberta (Bei Jiang)$119k
- University of Toronto, Mila / HEC Montréal and McGill (Ebrahim Bagheri, Jian Tang, Benjamin Fung)$72k
- Mila / Université de Montréal (Yoshua Bengio)$72k
All grants
13 grants totalling $1.1M, newest first. Recipients link to their own pages; people who received grants without a public profile are shown by number.
2026 4 grants · $475k
| Date | Recipient | Amount | Purpose |
|---|---|---|---|
| Feb 2026 | Vector Institute / University of Toronto (Zhijing Jin) | $119k | CAISI co-funding of an Alignment Project round 1 grant: Game-theoretic safety guarantees for advanced AI systems. CA$165,000 for one year (possible second year), at CA$1 = US$0.72; CIFAR's share only, the Alignment Project row has no amount source |
| Feb 2026 | Amii / University of Alberta (Linglong Kong) | $119k | CAISI co-funding of an Alignment Project round 1 grant: Sample-efficient online fine-tuning against resistant behaviors: statistical foundations for post-training alignment. CA$165,000 for one year (possible second year), at CA$1 = US$0.72; CIFAR's share only, the Alignment Proje source |
| Feb 2026 | University of Toronto (Yonatan Kahn) | $119k | CAISI co-funding of an Alignment Project round 1 grant: Scaling laws, data distributions, and learning dynamics: simulated high-energy physics data as a benchmark for data in the wild. CA$165,000 for one year (possible second year), at CA$1 = US$0.72; CIFAR's share only, the Alig source |
| Feb 2026 | Amii / University of Alberta (Bei Jiang) | $119k | CAISI co-funding of an Alignment Project round 1 grant: A unified statistical framework for quantifying rare event risks for language models. CA$165,000 for one year (possible second year), at CA$1 = US$0.72; CIFAR's share only, the Alignment Project row has no amount source |
2025 9 grants · $648k
| Date | Recipient | Amount | Purpose |
|---|---|---|---|
| Jun 2025 | University of Toronto, Mila / HEC Montréal and McGill (Ebrahim Bagheri, Jian Tang, Benjamin Fung) | $72k | CAISI AI Safety Catalyst Grant (first round): Adversarial robustness in knowledge graphs. CA$100,000 for one year, at CA$1 = US$0.72 source |
| Jun 2025 | Mila / Université de Montréal (Yoshua Bengio) | $72k | CAISI AI Safety Catalyst Grant (first round): Sampling latent explanations from LLMs for safe and interpretable reasoning. CA$100,000 for one year, at CA$1 = US$0.72 source |
| Jun 2025 | Amii / University of British Columbia (Mi Jung Park) | $72k | CAISI AI Safety Catalyst Grant (first round): On the safe use of diffusion-based foundation models. CA$100,000 for one year, at CA$1 = US$0.72 source |
| Jun 2025 | Vector Institute / Johns Hopkins (Gillian Hadfield) | $72k | CAISI AI Safety Catalyst Grant (first round): Advancing AI alignment through debate and shared normative reasoning. CA$100,000 for one year, at CA$1 = US$0.72 source |
| Jun 2025 | Mila / Université de Montréal (Gauthier Gidel) | $72k | CAISI AI Safety Catalyst Grant (first round): Adversarial robustness of LLM safety. CA$100,000 for one year, at CA$1 = US$0.72 source |
| Jun 2025 | Vector Institute / University of Toronto (Alán Aspuru-Guzik) | $72k | CAISI AI Safety Catalyst Grant (first round): Safe autonomous chemistry labs. CA$100,000 for one year, at CA$1 = US$0.72 source |
| Jun 2025 | Mila / Polytechnique Montréal and Amii / University of Alberta (Foutse Khomh, Lei Ma, Randy Goebel) | $72k | CAISI AI Safety Catalyst Grant (first round): Safety assurance and engineering for multimodal foundation model-enabled AI systems. CA$100,000 for one year, at CA$1 = US$0.72 source |
| Jun 2025 | Mila / McGill University (Jackie Chi Kit Cheung, Jin Guo) | $72k | CAISI AI Safety Catalyst Grant (first round): Maintaining meaningful control: Navigating agency and oversight in AI-assisted coding. CA$100,000 for one year, at CA$1 = US$0.72 source |
| Jun 2025 | Vector Institute / University of Toronto (Sheila McIlraith) | $72k | CAISI AI Safety Catalyst Grant (first round): Formalizing constraints for assessing and mitigating agentic risk. CA$100,000 for one year, at CA$1 = US$0.72 source |