Canada launched the Canadian AI Safety Institute (CAISI) in November 2024 with $50 million in funding—making it one of the first countries to establish a dedicated federal body for AI safety research and policy. Here's everything builders, researchers, and policymakers need to know.
What Is the Canadian AI Safety Institute?
The Canadian AI Safety Institute (CAISI) is Canada's federal organization dedicated to ensuring artificial intelligence is developed and deployed safely. Launched in November 2024 by Innovation Minister François-Philippe Champagne, CAISI operates as part of Canada's broader $2.4 billion AI investment strategy.
CAISI represents Canada's commitment to being a leader—not just a participant—in global AI safety governance.
"Canada has been a leader in AI since its inception. We must now lead in ensuring AI is developed responsibly and safely." — François-Philippe Champagne, Minister of Innovation, Science and Economic Development, November 2024
Why Did Canada Create CAISI?
Canada created CAISI in response to three converging pressures: rapid AI capability advancement, international momentum on AI safety, and Canada's unique position as home to AI pioneers like Geoffrey Hinton and Yoshua Bengio who have publicly warned about AI risks.
The Catalyst Events
Canada's Strategic Position
Canada occupies a unique position in global AI:
- Home to AI pioneers: Hinton (Toronto), Bengio (Montreal), and their students built modern deep learning
- Research infrastructure: Vector Institute, Mila, and Amii form the world's strongest public AI research network
- International credibility: Trusted by both US and European partners
- Talent pipeline: Thousands of AI researchers trained in Canadian universities
CAISI leverages this position to influence global AI safety standards while protecting Canadian interests.
Who Leads CAISI?
Executive Leadership
Dr. Elissa Strome was appointed as CAISI's first Executive Director in February 2025. Her background uniquely positions her to bridge research and policy.
Strome previously led CIFAR's Pan-Canadian AI Strategy—the $125 million initiative that created Vector, Mila, and Amii. She understands both the technical landscape and the policy machinery needed to govern AI effectively.
Research Program Leadership
CAISI's research program is managed through CIFAR with dedicated co-directors:
Nicolas Papernot is a leading researcher in AI security and privacy, known for work on adversarial machine learning. Catherine Régis brings expertise in AI governance and health law.
How Is CAISI Funded?
CAISI operates with a $50 million budget over five years, part of Canada's larger $2.4 billion AI investment announced in Budget 2024.
Budget Breakdown
Comparison to International AI Safety Bodies
Canada's funding is modest compared to the UK but represents serious commitment for a country of Canada's size.
What Research Does CAISI Fund?
CAISI funds AI safety research through multiple mechanisms, administered primarily through CIFAR's research program.
AI Safety Catalyst Grants (2025)
In June 2025, CAISI announced the AI Safety Catalyst Grants—$100,000 awards to accelerate promising safety research.
AI Safety Solution Networks (2025)
In November 2025, CAISI launched AI Safety Solution Networks—larger collaborative grants for multi-institution research.
Research Priority Areas
CAISI's research program focuses on:
How Does CAISI Connect Internationally?
CAISI is a founding member of the International Network of AI Safety Institutes—a coalition of national AI safety bodies coordinating on research and standards.
International Network Members
UK AI Security Institute Partnership
In 2025, CAISI pledged $1 million to the UK AI Security Institute's Alignment Project—a collaborative initiative on AI alignment research.
Yoshua Bengio's participation on the advisory board connects CAISI to Mila's safety expertise while giving Canada influence on international alignment research priorities.
"International cooperation on AI safety isn't optional—it's essential. AI systems don't respect borders, and neither can our safety efforts." — Common view among CAISI leadership
How Does CAISI Work with Vector, Mila, and Amii?
CAISI's research program operates through CIFAR, which coordinates with Canada's three national AI institutes.
Institute Collaboration Model
This structure ensures CAISI research benefits from Canada's existing AI infrastructure rather than duplicating it.
CIFAR's Role
CIFAR (Canadian Institute for Advanced Research) manages CAISI's $27 million research allocation:
- Grant administration: Distributes Catalyst Grants and Solution Networks funding
- Research coordination: Connects researchers across institutions
- International liaison: Maintains relationships with global AI safety community
- Talent development: Supports trainees in AI safety research
What Has CAISI Accomplished So Far?
Since launching in November 2024, CAISI has moved quickly to establish operations and fund research.
CAISI Timeline
Early Priorities
CAISI's first year has focused on:
- Building infrastructure — Hiring staff, establishing processes
- Launching research programs — Catalyst Grants, Solution Networks
- International engagement — UK partnership, network participation
- Stakeholder consultation — Engaging Canadian AI community
How Can Researchers Engage with CAISI?
Funding Opportunities
Engagement Pathways
For Academic Researchers:
- Apply for Catalyst Grants through CIFAR
- Propose Solution Network collaborations
- Connect through Vector, Mila, or Amii affiliations
For Industry:
- Participate in consultation processes
- Collaborate on safety evaluation methods
- Engage through industry association channels
For International Researchers:
- Connect through International Network member institutes
- Collaborate with Canadian partners on joint proposals
- Attend CAISI-sponsored workshops and events
How Does CAISI Compare to Other Safety Organizations?
Canadian AI Safety Landscape
Key Differences from US and UK Models
Canada's model emphasizes leveraging existing research infrastructure (Vector, Mila, Amii) rather than building new evaluation capacity from scratch.
What Are CAISI's Current Priorities?
Based on 2025-2026 announcements and activities:
Near-Term Focus (2025-2026)
Strategic Challenges
CAISI faces several challenges common to new AI safety bodies:
- Speed of AI development — Safety research struggles to keep pace with capabilities
- Talent competition — AI safety researchers are in high demand globally
- Coordination complexity — Balancing research independence with policy needs
- International alignment — Harmonizing approaches across different national bodies
What Does CAISI Mean for Canadian AI Builders?
For Startups
- Safety by design: CAISI resources can inform responsible AI development practices
- Competitive advantage: Demonstrating safety practices increasingly matters to customers and investors
- Regulatory anticipation: CAISI's work signals where Canadian AI regulation may head
For Researchers
- Funding opportunities: Catalyst Grants and Solution Networks provide dedicated safety funding
- Career paths: AI safety is becoming a recognized research specialization
- International connections: CAISI links Canadian researchers to global safety community
For Enterprise
- Best practices: CAISI outputs inform enterprise AI governance
- Evaluation frameworks: Upcoming CAISI evaluation methods will guide internal assessments
- Regulatory preparation: Understanding CAISI priorities helps prepare for future requirements
How Can You Stay Connected to CAISI?
Official Channels
- ISED announcements: Official CAISI news via Innovation Canada
- CIFAR updates: Research program news through CIFAR communications
- International Network: Global AI safety developments
Community Connections
- AGI House Canada: Join our AI safety discussions and events
- Vector/Mila/Amii events: Institute-hosted safety seminars and workshops
- Academic conferences: NeurIPS, ICML safety tracks feature CAISI-adjacent research
Related Reading
AI Safety Leaders
- Yoshua Bengio: Pioneer of Deep Learning — Turing Award winner leading AI safety
- Geoffrey Hinton: Godfather of AI — Nobel laureate warning about AI risks
- Ilya Sutskever & SSI — From OpenAI to Safe Superintelligence
AI Safety Hubs
- Montreal: Where AI Safety Meets Breakthrough Research — Canada's AI safety capital
- Mila Complete Guide — The world's largest academic AI research center
The Broader Canadian AI Ecosystem
- The AI Triangle: Canada's Three Superpowers — How Canadian AI is shaping global safety discussions
Join AGI House Canada to connect with Canada's AI safety community. Scan the QR code to join our WhatsApp group or subscribe to our newsletter.
Data current as of January 2026. Information from ISED announcements, CIFAR communications, and official government sources.
