Top Insights for AI’s Impacts on Trademark Practice in 2026
Alt Legal Team | January 22, 2026
This article contains insights derived from the Alt Legal Webinar, “Mark to the Future: Predictions for AI’s Impacts on Trademark Practice,” where Jing He (Partner at Gen Law in China) and Professor Daryl Lim (Penn State Dickinson Law) explored how AI is reshaping trademark practice in the U.S. and China and where the biggest opportunities and risks lie.
To better understand how AI is shaping the future of trademark law, here’s what our speakers predict for 2026 and beyond.
1. There are many helpful AI tools that positively impact on trademark practitioners’ workflows.
AI is being used as key tool in trademark practice, helping to:
- Generate presentations
- Conduct research (it’s a great search engine)
- Provide trademark naming and refusal risk assessment
- Deliver refusal review
- Draft client analysis opinion letters and IP office responses,
- Demonstrate evidence of use in non-use cancellation stage
- Conduct similarity analysis for search and clearance
- Review cross-border compliance.
Additionally, for attorneys working on international matters, it can help them write in foreign languages and polish their work. (Jing)
2. While AI can facilitate the trademark prosecution process, it cannot replace human judgment when it comes to likelihood of confusion.
Even though AI is providing practitioners with so many tools to make trademark search, clearance and risk assessment much easier and at scale, likelihood of confusion is still based on human judgement. AI can assess risk but it cannot assess why similarity matters or whether it is legally actionable. Likelihood of Confusion isn’t just a visual test, it’s market and consumer-specific. Put another way, AI sees pixels and courts see consumers. (Daryl)
3. AI is changing the way consumers make shopping decisions and ultimately how likelihood of confusion plays out in the marketplace.
Consumers have shifted their trust from brands to shopping platforms. Platforms are steering outcomes with algorithms and delivering the choices as opposed to consumers actively choosing a brand and searching for it. Confusion is no longer about two marks colliding, it’s about how online shopping platforms are steering attention with recommendations and outcomes based upon search terms. (Daryl)
4. AI has strengths and weaknesses when it comes to trademark search, clearance, and monitoring.
AI is a phenomenal tool for conducting these tasks at scale. It sees phonetic similarities across languages, recurring visual motifs, and repeated bad actors. However, AI struggles with meaning. AI cannot identify whether there is similarity or confusion in a way that is relevant to trademark law. This is where practitioners need to step in and decide what matters. Remember that trademark law doesn’t punish similarity, it punishes confusion – there is a legal difference. (Daryl)
5. AI is a helpful tool for conducting research on international trademark law.
AI is giving practitioners more confidence when handling international trademark issues. Before AI, practitioners relied heavily on global colleagues to advise on international matters. AI is giving us the confidence to handle some of this initial research so that when we work with our foreign colleagues, we are providing more effective service. We aren’t replacing foreign colleagues, but rather creating higher-demand for them. (Jing)
6. Avoid over reliance on AI for trademark clearance.
Even if an AI tool cleared a mark, courts will never consider this as evidence. Courts will always look at whether there is likelihood of confusion from a consumer perspective. When using AI tools for clearance, keep in mind that these tools can lower perceived risk without lowering actual legal exposure. Using AI in this context deserves the same scrutiny applied to junior attorney work; trust the outcome, but verify. (Daryl)
7. Consult others with AI outputs.
We know not to rely too heavily on AI outputs. Talk to senior lawyers and use group intelligence to gather more informed judgment based on AI outputs. Don’t rely on a single person’s work and watch out if things get too easy with AI. (Jing)
8. Courts may become more skeptical about AI-produced findings.
AI won’t replace judicial judgment but it will raise transparency and methodology, just like survey evidence faces a high level of scrutiny. Even if we acknowledge that we used AI to come to certain conclusions, courts may question the AI itself. Courts may start to ask – How did AI come to that conclusion? What was the data used to train AI? Can the AI training process be audited? (Daryl)
9. Use caution when creating brand assets with AI as these models are trained on existing brands and trademarks.
Watch this space carefully and think about the overlap with trademark naming and copyright law. Trademark law is agnostic when it comes to who came up with the trademark whereas that is not the case in copyright or patent law. Trademark law does not reward creativity, rather you must still show use in commerce and distinctiveness. (Daryl)
10. AI can be a powerful tool for detecting counterfeits, but it can also produce counterfeits very effectively.
Service providers have developed tools to review counterfeits and AI makes this a lot easier to detect. (Jing) On the other hand, AI can make better counterfeits by producing deepfakes, fake domains, phishing scams, and small variations to evade detection. (Daryl)