Much of Canada's visible AI research runs through one federal program. The Pan-Canadian Artificial Intelligence Strategy, launched in 2017 with CAD 125 million in Budget 2017 and administered by CIFAR, funds three national institutes (Amii in Edmonton, Mila in Montreal, and the Vector Institute in Toronto) and the Canada CIFAR AI Chairs, a group of more than 125 researchers. One of those chairs, Richard Sutton, shared the 2024 ACM A.M. Turing Award for work on reinforcement learning.
Key takeaways
- The first phase of the strategy (2017 to 2022) received CAD 125 million; the second phase (2021 to 2031) added CAD 443.8 million in Budget 2021 and CAD 174 million in the 2024 Fall Economic Statement.
- CIFAR named 29 inaugural Canada CIFAR AI Chairs on December 3, 2018, with a stated target of more than 50 leading researchers.
- Sutton, a University of Alberta professor and chief scientific advisor at Amii, shared the 2024 Turing Award with Andrew Barto.
- Reinforcement learning is learning from reward signals over time, and temporal difference learning is one of its best-known methods.
- Funding programs and prizes describe a researcher's standing, but they do not replace reading the papers.
How the national network is organized
According to Innovation, Science and Economic Development Canada, the strategy's first phase ran from 2017 to 2022 and had three components: the Canada CIFAR AI Chairs program, the three national AI institutes, and CIFAR-led research, training, and collaboration programs. CIFAR describes the strategy as the first national AI strategy in the world. Each chair is nominated by and affiliated with one of the institutes, which are located in Edmonton, Montreal, and Toronto.
The second phase, covering 2021 to 2031, shifted emphasis toward adoption. The table below lists the amounts the government page gives, in Canadian dollars. The page itself does not state a total or a status label for either phase.
| Item | Amount | Source of funds and period |
|---|---|---|
| First phase, whole strategy | CAD 125 million | Budget 2017, 2017 to 2022 |
| Second phase, first allocation | CAD 443.8 million | Budget 2021, 2021 to 2031 |
| Second phase, added allocation | CAD 174 million | 2024 Fall Economic Statement |
| National AI institutes | CAD 60 million | Budget 2021, up to CAD 20 million each through 2025-26, plus CAD 24 million over three years from 2025-26 |
| CIFAR, including the Chairs | CAD 208 million | Budget 2021, 2021-22 to 2030-31 |
| National AI compute | CAD 40 million | Budget 2021, 2022-23 to 2026-27, through the Digital Research Alliance of Canada |
The Canada CIFAR AI Chairs
CIFAR announced the first cohort of 29 chairs on December 3, 2018, at AICan, the first annual meeting of the strategy, and reported about CAD 30 million in funding announced that day. An International Scientific Advisory Committee, drawing on people from industry labs and universities, reviewed nominations from the institutes. CIFAR's announcement describes the chairs' work as covering areas such as natural language processing, image recognition, and reinforcement learning. Its current program page says more than 125 researchers are supported, with research that now also spans AI safety, drug discovery, machine learning for health, and materials discovery.
CIFAR also states that the chief science advisors of all three national institutes are Turing Award winners. Sutton is Amii's chief scientific advisor and has been a Canada CIFAR AI Chair since 2021.
One researcher's work: reinforcement learning
ACM announced the 2024 Turing Award in March 2025, and the University of Alberta's release credits Sutton and Barto with pioneering reinforcement learning. Their textbook, Reinforcement Learning: An Introduction, appeared in 1998, and CIFAR's announcement names temporal difference learning among their most prominent contributions. The award carries a USD 1 million prize, funded with support from Google, according to the university.
In reinforcement learning, a program learns by trying actions and receiving rewards, instead of being shown correct answers. Temporal difference learning adjusts an estimate using the next estimate. The numbers here are illustrative. Suppose a game position A is currently valued at 0.5 and a move leads to position B valued at 0.8, with no immediate reward. With a step size of 0.1 and no discounting, the new value of A is 0.5 + 0.1 x (0 + 0.8 - 0.5) = 0.53. A moves a little toward B, without waiting for the game to end. Repeated over many games, estimates spread backward from the outcomes.
How to read a researcher's profile
- Check the primary record: the papers themselves, not only summaries of them.
- Separate the prize citation from the full history. Committees name a few people, and the fields usually involve many more.
- Note the affiliation and funding role, such as a chair or an institute position, because it shows where the work is supported but not whether it is correct.
- Look for replication or follow-up work by independent groups.
- Treat rankings of the "most renowned" researchers with caution, since no neutral, agreed method produces them.
FAQ
What is CIFAR?
CIFAR is a Canadian research organization that the federal government chose in 2017 to lead the Pan-Canadian Artificial Intelligence Strategy. It runs the AI Chairs program and national research and training programs.
Where are Canada's national AI institutes?
Amii is in Edmonton, Mila is in Montreal, and the Vector Institute is in Toronto.
What is reinforcement learning in plain terms?
It is a way of training a program through trial, error, and reward signals. The program is not given labeled examples; it learns which actions tend to lead to better outcomes.
Does a Turing Award mean a researcher's every claim is correct?
No. The award recognizes specific contributions as judged by a committee. Claims in later papers or interviews still need to be checked on their own evidence.
Related reading
- more on leading AI researchers
- how deep learning works
- AI and the future of work
- examples of creative AI
More in AI Basics.
Last verified: October 2026
Sources
- Pan-Canadian Artificial Intelligence Strategy, Innovation, Science and Economic Development Canada, 2026
- 29 researchers named to first cohort of Canada CIFAR Artificial Intelligence Chairs, CIFAR, December 3, 2018
- Canada CIFAR AI Chairs and national AI institutes overview, CIFAR, 2026
- Rich Sutton honoured with Turing Award for major contributions to AI, CIFAR, March 10, 2025
- U of A computing science professor and Amii fellow wins Nobel Prize in computing, University of Alberta, March 2025