AI Is Rewriting the Rules for Creating and Protecting Trademarks

The sheer volume of AI adoption has prompted wide-scale changes in how trademarks are created, cleared, challenged, and enforced. The International Trademark Association in March reported that 74% of in-house practitioners surveyed now actively use artificial intelligence tools.

For both general counsel and outside counsel, this is redefining trademark strategy in concrete ways. Clients expect advice that reflects how AI is affecting examiner behavior, adversary tactics, and marketplace dynamics.

Below are five areas where GCs and outside counsel should reconsider their approach.

Creative Pipeline Altered

A marketing team can now produce dozens of names and logos in a single session with a generative tool. But that speed comes with risks. Many AI models are trained on datasets that include existing brands, so a “new” asset may unintentionally resemble someone else’s mark. At the same time, AI-generated names and logos tend to produce similar styles, making it harder to develop something truly distinctive.

For GCs, this is operational issue as well as a legal one. If a designer uses an AI tool without involving legal counsel, the team may have no visibility into whether the output replicates existing marks. Documenting meaningful human creative decisions is essential, including what the team contributed and how they directed the output, not just whether AI was used. That record matters not only for registrability but also for preserving enforcement rights later.

Less Reliable Results

The speed of AI-assisted clearance tools presents another major issue: a false sense of confidence. While effective at spotting obvious visual matches, they often miss nuances such as phonetic similarity, marketplace context, or consumer perception. They struggle with edge cases, including foreign equivalents, stylized marks, and niche terminology.

AI can recognize patterns, but it can’t account for how real consumers think or react. That distinction matters because trademark law depends on consumer perception, and an algorithm can’t run a true likelihood-of-confusion analysis.
For GCs and outside counsel, AI can identify potential issues, but its output should never be the final determination. Human judgment should still be the deciding factor in close calls.

Enforcement Arms Race

AI is being used by examiners, adversaries, and infringers alike. The USPTO’s Class ACT tool, which automatically assigns classes, design search codes, and pseudo-marks, launched in March and marks a turning point in examination. Applicants now need to consider how an AI system, not just a human examiner, might evaluate a mark during prosecution.

On the enforcement side, economics have shifted. Infringers who once ignored demand letters or had to hire a lawyer can now generate professional responses quickly using AI tools. Opposers and petitioners are using AI to draft arguments and identify prior rights more efficiently.

Courts are seeing AI-generated content cited as fact, including hallucinated citations and fabricated specimens. Because AI outputs can’t be fully explained, it can be difficult to determine whether elements in a filing were intentional or a byproduct of training data. That uncertainty makes disputes more complex, costlier, and less predictable.

The result is a shift in how practitioners approach filings and disputes. It’s worth assuming that AI tools will be reviewing submissions at some stage, and it’s equally important to double-check citations and supporting materials, even when they appear accurate at first glance.

Counterfeiting, Marketplace Abuse

AI has made it inexpensive to produce convincing knockoffs. A single infringer can generate dozens of variations in minutes. At the same time, as “dupe culture” grows, products start to mimic a brand’s look and feel without clearly infringing.

Tracking individual listings no longer is effective. Brand owners are pivoting to strategies that track patterns across platforms. They are moving faster to keep up with infringing products using real-time detection tools that monitor live marketplaces. The Model Context Protocol, a standard that allows AI systems to query external databases in real time, eventually could enable AI tools to check trademark records and marketplace listings as part of a unified workflow.

Even with these tools, human oversight is critical. Someone must verify the findings, assess the risk, and decide which action makes sense. Brands that succeed will spot trends early and act before problems escalate.

READ MORE: IP vs. AI: A New Frontier of Legal Battles to Protect Creativity

The Inside Risk

While monitoring external threats is crucial, internal use of AI also presents serious risks. Employees often use generative tools to brainstorm names, draft content, and create visual assets without considering trademark implications. A prompt entered into a public model could reveal confidential plans, which may then remain stored in AI models beyond the company’s control.

This is a governance issue that belongs on legal’s agenda. Companies need clear, AI-specific brand policies: which tools are allowed, what must be disclosed to legal, and who can approve AI-generated assets. They must audit how their marks appear in external datasets and whether opt-out mechanisms or contractual protections are appropriate.

Without clear guardrails, everyday use of AI can create ongoing issues that are difficult to reverse and may expose competitive information.

Looking Ahead

For general counsel, the immediate priority is updating systems, workflows, and agreements that govern how marks are created, cleared, protected, and enforced. Outside counsel need to understand how AI is influencing examiner decisions, adversary behavior, and marketplace dynamics, so clients can adjust before a dispute arises.

The goal is to adopt a policy and ensure legal teams are involved early enough that even a simple prompt doesn’t create problems now or in the future.

This article does not necessarily reflect the opinion of Bloomberg Industry Group, Inc., the publisher of Bloomberg Law, Bloomberg Tax, and Bloomberg Government, or its owners.

Author Information

Barry Greenbaum is a partner with Olshan’s brand management and protection and intellectual property law practices.

Write for Us: Author Guidelines