Canada’s AI rules expand through laws, courts and regulators

The University of Windsor Law Library’s Canada tracker organizes federal, provincial and territorial material on AI regulation, including legislation, bills, policies, court practice directions, judicial decisions, investigations, standards, reports, benchmarks and international cooperation. At the federal level, it highlights Bill C-27, which would have enacted privacy and AI legislation but was not enacted at dissolution of Parliament, alongside government guidance on automated decision-making, generative AI and responsible development of advanced systems.

Provincial entries point to a widening mix of sector-specific rules and policy tools. Ontario’s public sector measures address cyber security, AI accountability, risk management and oversight, while employment standards require disclosure when AI is used to screen, assess or select applicants. Quebec materials include privacy modernization, public-body guidance on generative AI and responsible deployment work, while law societies across Canada have issued practice resources for lawyers using generative AI.

Courts and tribunals have become a central focus, with practice directions requiring parties to disclose, verify or take responsibility for AI-assisted filings. The collected decisions show repeated warnings over fabricated citations, unsupported legal propositions and AI-generated evidence, with consequences ranging from disregarded submissions to costs awards, sanctions and referrals to law societies. Other entries cover privacy investigations involving ChatGPT and X, Canadian media litigation against OpenAI, deepfake proceedings, exam proctoring, patent inventorship and international AI cooperation statements.