Overview
Current Canadian copyright law faces a growing challenge in addressing creative works generated or assisted by artificial intelligence (AI). The law remains rooted in traditional notions of human authorship, leaving a gap between technological capability and legal recognition. While the United States, United Kingdom, and other jurisdictions have begun to develop guidance and case law in this area, Canada lags, creating uncertainty in authorship, ownership, and enforceability—particularly for those working in film, television, music, and digital media.
1. Current Legal Landscape
- No legislative framework currently addresses AI-generated works under Canadian copyright law. • Copyright protection requires a human author; works produced autonomously by AI are not protected.
- Other jurisdictions (notably the U.S.) are further ahead, with administrative and court decisions clarifying limits on AI authorship.
- The absence of Canadian precedent leaves both creators and users of AI-generated works in a state of legal uncertainty.
2. Human Authorship Requirement
Canadian copyright law presumes authorship arises from a natural person. For AI-assisted works, the determining factor is the degree of human creativity and control. U.S. developments (e.g., Thaler v. Perlmutter) reaffirm that human authorship is essential, that AI may serve as a tool but cannot itself be the author. Even detailed or complex AI prompts may not meet the legal threshold for human authorship if the creative decision-making is primarily machine-driven. The guiding analogy: AI is “playdough in human hands” with the outcome depending on how much the human shapes the result.
3. Legal Precedents, Liability, and Contractual Protections
Older Canadian precedents are no longer reliable in this emerging context. However, the matter of CIPPIC v. Sahni may eventually offer some clarity once the courts have decided the case.
For now, contractual terms are the primary mechanism for managing ownership and risk:
- Include representations and warranties around originality and authorship.
- Clearly allocate liability where AI tools are involved in content creation.
- Maintain records of human input, creative control, and revisions.
Performers must protect image, likeness, and licensing rights because AI can replicate or simulate human appearances and voices.
4. Litigation and Enforcement Challenges
Determining which portions of an AI-generated work carry human authorship, and thus protection, is complex and untested. Litigation is expected to be expensive and fact-specific, requiring expert evidence about how AI was used and to what extent humans guided the output. Documentation and transparency in creative workflows are crucial to support claims of human authorship.
5. CRTC and Canadian Content Implications
The CRTC’s current “Canadian content” (CanCon) point system focuses on human creative roles (writers, directors, performers). AI participation is not recognized within existing assessment frameworks.
Suggested policy directions:
- Reduce emphasis on IP ownership as the sole criterion for CanCon recognition.
- Consider alternative metrics (e.g., location of production, Canadian supervision, cultural intent).
- Explore how AI could enhance Canadian creative output rather than exclude it.
6. Cross-Border and Jurisdictional Issues
Jurisdictional questions are emerging:
- What happens when a Canadian company uses AI tools hosted abroad?
- Can a Canadian entity claim authorship of a work generated by AI operating outside Canada?
- Conversely, if a foreign creator uses Canadian AI infrastructure, could the resulting work qualify as Canadian content?
These scenarios highlight both opportunities for cross-border collaboration and risks arising from legal ambiguity.
7. Practical Strategies and Risk Mitigation
Given the uncertainty, practitioners should focus on contractual and procedural safeguards:
- Define ownership and authorship expressly in contracts.
- Register works and maintain detailed records of human contribution.
- Include robust representations, warranties, and indemnities.
- Establish chain of title early; clarify ownership before licensing or financing.
8. The Value of Clarity
Intellectual property uncertainty can delay deals, impede financing, and diminish IP valuation. Proactive documentation and precise agreements can preserve asset value and ensure smoother transactions. Legal and operational clarity remains the most practical safeguard in the current legislative vacuum.
9. Policy Outlook and Recommendations
The OBA is well-positioned to contribute to national discussions on modernizing copyright law to reflect AI’s role in creation. Policy goals should include:
- Recognizing AI-assisted authorship where meaningful human creativity exists.
- Clarifying ownership and moral rights in hybrid human–AI works.
- Integrating AI considerations into Canadian content policy.
Continued dialogue between lawyers, creators, policymakers, and regulators is critical as Canada moves toward a workable framework.
Key Takeaways
- Canada lacks a defined legal framework for AI-generated works, resulting in uncertainty.
- Human authorship remains the cornerstone of protection; AI cannot hold copyright.
- Contracts, documentation, and careful licensing are essential mitigation tools.
- CRTC and cultural policy frameworks must evolve to account for AI’s growing role in creative production.
- The OBA has an opportunity to lead the policy conversation and advocate for legislative modernization.
- Prepared by Edward Peghin for the OBA
Any article or other information or content expressed or made available in this Section is that of the respective author(s) and not of the OBA.