7 Key Takeaways: The Current AI Landscape in Trademarks and Copyright
Kilpatrick’s Joe Petersen and Briggs Wright recently presented “The Current AI Landscape in Trademarks and Copyright” at the firm’s annual Advanced Trademark Law Seminar in New York.
1. Use AI to Drive Efficiency — Not to Replace Judgment
Internal legal teams and outside counsel are broadly embracing AI, primarily as an efficiency and workflow tool rather than a substitute for attorney judgment or final work product. Common uses include:
- Summarizing lengthy documents, legal materials, and complex email chains into concise, business-friendly formats.
- Generating first-pass outlines or draft language for contracts, policies, and training materials.
- Stress-testing arguments by using AI as a brainstorming partner to surface gaps, counterarguments, or alternative perspectives.
The consistent message: AI is most valuable upstream in the process, not as the final author.
2. Transparency and Billing Expectations Are Rising
Clients generally do not prohibit AI use by outside counsel, but expectations around transparency are increasing:
- Disclosure: Some clients now request that AI use be flagged in billing entries, particularly where it may reduce time spent on junior-level research or drafting.
- Human Review Is Mandatory: There is no tolerance for unreviewed AI output. All AI-assisted work must be independently evaluated by an attorney for accuracy, legal reasoning, and completeness.
AI can accelerate work, but responsibility remains squarely with counsel.
3. Human Contribution Remains Central to IP Protection
For branding, marketing, and creative initiatives, human involvement is critical to preserving intellectual-property rights:
- Copyrightability: Many organizations now require that AI outputs be grounded in original, human-created material to strengthen claims of protectability.
- Vendor Controls: AI-specific addenda are increasingly common in vendor agreements, including requirements to disclose prompts and confirm that proprietary data was not misused.
- Clearance Still Matters: Speed does not eliminate risk. Traditional clearance and human-led review remain essential for new brands, logos, and creative assets to ensure enforceability.
Documenting human creative input continues to be critical.
4. Enterprise-Grade AI Is Becoming the Norm
To protect privileged and confidential information, companies are moving decisively toward secure, enterprise-grade AI platforms:
- Private Tenants: These environments prevent internal data from being used to train public or general models.
- Public Tool Restrictions: Some organizations rely on warnings and usage guidelines, while others have blocked public AI tools entirely on company devices to reduce the risk of accidental disclosure.
Data security and privilege protection are driving technology choices as much as functionality.
5. Expect New Legal Risks — and More Litigation, Not Less
AI adoption is reshaping the litigation landscape in ways that may increase, rather than reduce, legal workloads:
- Regulatory Pressure: Emerging global laws addressing data transparency, attribution, and remuneration are expected to be a major focus over the next 12–24 months.
- Lower Barriers to Entry: AI is enabling less-sophisticated complainants to generate more polished complaints, potentially increasing the volume of disputes.
- Process as Protection: In copyright matters, the ability to show thoughtful guardrails, adverse prompting, and other risk-mitigation measures may be critical to defending fair-use and non-infringement arguments.
Strong governance and documentation around AI use are becoming as important as the technology itself.
6. The Legal Landscape for AI Training Is Rapidly Evolving
Federal courts are actively grappling with questions around the use of copyrighted materials for AI training, especially in major jurisdictions such as the Southern District of New York, Northern District of California, and District of Delaware.
7. Legislative and Regulatory Responses Are Accelerating
Governments are responding with new legislation and regulatory guidance tailored to AI. Notable U.S. bills include the TAKE IT DOWN Act (signed into law), the NO FAKES Act, the TRAIN Act, and the No Adversarial AI Act, reflecting Congressional attention to AI’s impact on copyright and content authenticity. In the EU, the AI Act imposes strict requirements for labeling synthetic content and compliance with copyright laws, with significant penalties for violations — up to €35 million or 7% of annual sales.
The U.S. Copyright Office’s guidance underscores the continued need for human authorship, recognizes market dilution as a form of harm, but also signals that existing copyright laws remain largely adequate for addressing these new challenges.
