Canada GenAI Code of ConductVoluntaryCanada (federal)

Voluntary Code of Conduct on the Responsible Development and Management of Advanced Generative AI Systems

Last reviewed Sep 25, 2026.

In short

Launched in September 2023 while AIDA was still before Parliament, the Code sets out six outcomes that signatories commit to for advanced generative AI: accountability, safety, fairness and equity, transparency, human oversight and monitoring, and validity and robustness. Developers commit to risk assessments, red-teaming, content provenance measures and publishing capabilities and limits; managers commit to monitoring after deployment. With AIDA dead, the Code is currently Canada's main federal statement of expected practice for generative AI, and ISED lists around 50 signatories.

Who it applies to

Organisations that build or use AI systems in Canada (federal).
Use cases in scope: generated content, chatbots, general use.
Size: everyone, including solo operators.
Voluntary. Aimed at developers and managers of advanced generative AI systems; signing signals good practice to customers and regulators but creates no legal duty.

Key dates

Sep 27, 2023✓Launched by Innovation, Science and Economic Development Canada
May 27, 2024✓Eight more organisations sign (30 signatories at the time)

What you have to do

Assess and mitigate risks (including bias and misuse) before deploying an advanced generative system.
Publish a plain-language description of the system's capabilities, limitations and intended use.
Use watermarking or provenance measures so AI-generated content can be identified.
Keep humans monitoring the system after deployment and report serious incidents.
Test for robustness and security, including adversarial (red-team) testing.

Penalties

None. Voluntary commitments; reputational consequences only.

Related lessons

Level 1The world map of AI rules: EU, US patchwork, Canada, standards8 min
Level 4Training and AI literacy programmes6 min

Real cases

Ruling2026 · CA-federal
Canadian privacy regulators v. OpenAI — joint findings on ChatGPT training data
Complaint well-founded. The federal OPC treated it as conditionally resolved on the strength of OpenAI's commitments (filtering tools, clearer notices about accuracy and sources, improved access tools, quarterly reporting); the Quebec, BC and Alberta offices found the consent and retention issues unresolved. No fine was imposed — PIPEDA gives the OPC no penalty powers.

Industries where it matters

Sources

Last reviewed Sep 25, 2026.