Artificial Intelligence · Canada
AI regulation in Canada: laws & policy (2026)
Canada shaded by its artificial intelligence status
AI in Canada: sectoral rules.
FrameworkPatchwork approach: federal Directive on Automated Decision-Making (Treasury Board, 2019); sectoral rules (OSFI Guideline E-23 for financial institutions, effective May 2027); the Voluntary Code of Conduct on Advanced Generative AI Systems (ISED, 2023); existing privacy laws (PIPEDA, Quebec Law 25); plus four AI-related bills tabled in 2026 by AI & Digital Innovation Minister Evan Solomon.
As of August 2026, Canada has no comprehensive federal AI law in force. The proposed Artificial Intelligence and Data Act (AIDA), part of Bill C-27, died on the Order Paper when Parliament was prorogued in January 2025 and Minister Evan Solomon has confirmed it will not be reintroduced. Canada currently regulates AI through a mix of sectoral instruments (OSFI Guideline E-23, the federal Directive on Automated Decision-Making), general laws (privacy, human rights, Criminal Code), the voluntary code of conduct for generative AI, and a new National AI Strategy launched June 4, 2026, with fresh AI-related bills tabled but not yet enacted.
Key points
The Artificial Intelligence and Data Act, introduced in June 2022 as part of Bill C-27, died on the Order Paper when Parliament was prorogued in January 2025 ahead of the federal election; the government has said any future AI law will be a new design, not an AIDA revival.
ISED launched a Voluntary Code of Conduct on the Responsible Development and Management of Advanced Generative AI Systems in September 2023; it is entirely voluntary with no enforcement body, penalties, or audits, and serves as an interim standard.
The Treasury Board's Directive on Automated Decision-Making (in force since April 1, 2019, compliance from April 1, 2020) is a mandatory, risk-based instrument governing public-facing automated decisions by ~97 federal institutions, backed by the Algorithmic Impact Assessment tool.
OSFI's final Guideline E-23 on Model Risk Management (issued 2025) applies to all federally regulated financial institutions and explicitly covers AI/ML models; it comes into force May 2027 after an 18-month transition and establishes a principles-based framework across the model lifecycle.
Prime Minister Mark Carney launched Canada's National Artificial Intelligence Strategy: AI for All on June 4, 2026, emphasizing trust, sovereign compute (a CAD $2B Sovereign AI Compute Strategy from Budget 2024), and international AI cooperation agreements with Germany, Norway and Finland.
AI Minister Evan Solomon has tabled a package of four bills in 2026, including an update to private-sector privacy law, legislation to regulate social media and AI chatbot companies, a ban on deepfakes in elections, and a criminal offence for non-consensual sexualized deepfakes; these are proposed and not yet in force.
Timeline - major decisions & events
ISED released 'Engagements on Canada's Next AI Strategy: Summary of Inputs,' synthesizing over 11,000 public submissions and 32 Task Force reports. The themes, risk-based regulation, sovereign infrastructure, talent, and liability, will shape the AI strategy expected to launch in 2026.
Innovation, Science and Economic Development Canada ↗Federal, provincial and territorial privacy authorities released a joint statement on AI-generated imagery (including deepfakes) and the protection of privacy, applying existing privacy law to synthetic media. Signals continued regulator-led governance of AI in the absence of a dedicated AI statute.
Office of the Privacy Commissioner of Canada ↗Minister Evan Solomon launched a 30-day national public engagement (Oct 1-31, 2025) and a 28-member AI Strategy Task Force to shape Canada's renewed AI strategy. It marked a pivot from the failed AIDA toward a new, adoption-focused governance framework.
Government of Canada ↗When Parliament was prorogued, Bill C-27, containing the Artificial Intelligence and Data Act (AIDA), Canada's would-be first comprehensive AI law, died on the Order Paper without passing. Canada was left with no dedicated federal AI statute.
Parliament of Canada (LEGISinfo) ↗Federal, provincial and territorial privacy commissioners issued joint principles for responsible, trustworthy and privacy-protective generative AI, mapping existing privacy law (legal authority, consent, transparency) onto AI development and use.
Office of the Privacy Commissioner of Canada ↗ISED Minister François-Philippe Champagne launched a voluntary code of conduct setting commitments on accountability, safety, fairness, transparency and human oversight for advanced generative AI. It was meant as an interim bridge until AIDA came into force.
Government of Canada ↗The Privacy Commissioner concluded the RCMP's use of Clearview AI's facial-recognition technology violated the federal Privacy Act, reinforcing limits on government use of AI biometric surveillance tools.
Office of the Privacy Commissioner of Canada ↗A joint investigation by federal and three provincial privacy commissioners found Clearview AI's scraping of billions of facial images was unlawful mass surveillance violating Canadians' privacy. It became a landmark application of privacy law to AI systems.
Office of the Privacy Commissioner of Canada ↗The federal government's first binding AI governance instrument required departments using automated decision systems to conduct Algorithmic Impact Assessments and meet transparency, accountability and fairness requirements. It remains Canada's main operative AI rulebook for government.
Treasury Board of Canada Secretariat ↗Budget 2017 committed $125 million to the world's first national AI strategy, funding research institutes (Amii, Mila, Vector) and talent. It established Canada's research-and-talent-first posture that still frames its 'enable innovation' approach to regulation.
Innovation, Science and Economic Development Canada ↗Canada - other topics
Artificial Intelligence in other countries
Last verified 8/16/2026 · Orientation, not legal advice - verify against the primary sources linked above. Methodology & how to cite · State of Technology Regulation 2026 · Explore the full world map →