Canadian AI Safety Institute

The Canadian Artificial Intelligence Safety Institute (CAISI) is Canada's national AI safety body, launched on November 12, 2024 in Montreal by the Honourable François-Philippe Champagne, Minister of Innovation, Science and Industry. CAISI was established with an initial budget of CAD $50 million over five years, funded through Canada's Budget 2024 as part of a broader $2.4 billion federal AI investment. The institute is administratively housed at Innovation, Science and Economic Development Canada (ISED) and operates with a dedicated office in Ottawa.

In February 2025, Dr. Elissa Strome was appointed as CAISI's first Executive Director. Dr. Strome previously served as Executive Director of the Pan-Canadian AI Strategy at CIFAR, bringing deep familiarity with Canada's AI research ecosystem. The institute's research is governed by a CAISI Research Council led by two co-directors: Nicolas Papernot (Canada CIFAR AI Chair, Vector Institute) and Catherine Régis (Canada CIFAR AI Chair, Mila).

CAISI conducts work through two parallel streams. The first is an applied and investigator-led research program administered by CIFAR (receiving $27 million of the total budget), which funds Solution Networks in AI Safety, AI Safety Catalyst Grants, and AI Alignment Projects. In its first year, the CIFAR program engaged a network of over 55 researchers and experts. The second stream consists of government-directed projects implemented by the National Research Council of Canada, focusing on AI evaluation frameworks, joint international testing exercises, and guidance for AI developers and the public.

CAISI is a founding member of the International Network of AI Safety Institutes, announced at the AI Seoul Summit in May 2024. NRC researchers have conducted joint testing exercises with counterparts from eight other countries, evaluating autonomous AI agents, multilingual content moderation, and AI performance on cybersecurity tasks. In 2025, CAISI also partnered with the United Kingdom to invest in AI alignment research. The institute also co-funded over $1 million in awards addressing sociotechnical AI safety challenges and a global AI alignment initiative.