IP: Trademarks, patents, & trade secrets

Introduction

As with copyright, trademark and patent law raise issues in relation to AI, and in particular generative AI. Trademarks play a crucial role in protecting a company’s identity and reputation, and issues are likely to arise in relation to AI-generated or AI-assisted art. In the context of AI, patents are used to protect technical inventions, including those developed with the aid of AI. However, specific problems can arise when AI is involved in the invention process. Finally, questions arise as to the scope and limits of trade secret protection in relation to AI.

Trademark and generative AI tools

We expect internal marketing departments to increasingly rely on generative AI to prepare creative content, which may be protected by trademark law, whether or not it is also protectable under copyright law. For example, a generative AI system might be asked to produce a slate of potential new product names, a fresh look for a webpage, a new slogan for an ad campaign, or a short audio signature or jingle for use when consumers interact with a new product or game. It’s worth remembering that any of these might be protectable by trademark law because they could serve as source indicators for consumers. Trademarks aren’t limited to company or product names; they can also include slogans, sound signatures (for example, the MGM lion’s roar), packaging designs, and more. When AI is used to generate these signatures, trademark clearance will be even more critical.

Where previously your internal marketing team might intuitively recognize a slogan or sound as already trademarked and steer clear of it, an AI-generated trademark might be just different enough not to trigger alarm bells during human review. Models trained on trademarked content, however, could generate outputs that infringe existing trademark rights. Trademark clearance, which is already advisable for all new brand indicia, will be especially critical for AI-generated or AI-assisted content. A robust clearance process will provide reassurance that whatever the output of an AI tool looks, reads, or sounds like, it will be compared against the trademark register to identify potentially conflicting marks before they become a problem in the marketplace. Trademark clearance provides a risk assessment of using the newly generated source indicator, enabling you to move your brand forward with a better understanding of the legal risks.

Trademark issues can also arise within copyrightable entertainment content. Should this occur, we encourage you to reach out to us to evaluate the use and whether it risks infringement or qualifies as fair use. It’s worth noting the Ninth Circuit decision in ESS Entertainment, where the Grand Theft Auto video game depicted a satirized version of the Play Pen club, and the club sued the game maker for trademark infringement.¹ In ESS Entertainment, the court found on the facts before it that the use was not explicitly misleading and was protected by the First Amendment. Issues like those in ESS Entertainment may arise in the context of AI-generated or AI-assisted art, where each element of a video game, movie, or commercial might not receive the same treatment it would otherwise receive if a human were responsible for every aspect of the design.

Patents and AI

Using generative AI to develop products or inventions for patenting presents both opportunities and risks. Generative AI promises to accelerate the development of inventions that benefit society, such as life-saving medicines, and there is an argument that AI should be recognized as an inventor on patents for such inventions. However, because AI is not human, it cannot be an inventor under the patent statutes of most countries. Using generative AI to develop products also creates a risk of patent infringement because the data used to train generative AI models may include patents or patented functionality.

On the issue of whether AI can be a named patent inventor, a growing consensus is clear: AI cannot be a named inventor, but human inventors who use AI as a tool in the inventive process can obtain patent protection for AI-assisted inventions as long as they can demonstrate a significant human contribution to the conception of the invention. In the United States, the Federal Circuit’s Thaler v. Vidal, 43 F.4th 1207, 1210 (Fed. Cir. 2022), affirmed a lower court’s ruling upholding the United States Patent and Trademark Office’s (USPTO) decision to deny petitions to name an AI system called Device for Autonomous Bootstrapping of Unified Sentience (DABUS) as a patent inventor. Based on U.S. Supreme Court precedent and language in the U.S. Patent Act, the Federal Circuit affirmed the holding that an inventor must be a natural person. However, the court left open the possibility that AI could contribute to a patented invention, stating that it was not addressing “the question of whether inventions made by human beings with the assistance of AI are eligible for patent protection.” In Thaler v. Comptroller-General of Patents, Designs and Trade Marks, the UK Supreme Court similarly held that UK patent legislation does not permit an AI system to be named as an inventor on a patent application.

Since this guide was first published, the USPTO has moved from soliciting input to issuing multiple rounds of formal guidance. Its current guidance anchors inventorship solely in traditional human conception, without establishing any separate AI-specific inventorship standard. Under that guidance, a natural person who uses an AI system as a tool in the inventive process can qualify as an inventor, provided that person makes a significant contribution to the conception of the invention. Merely prompting an AI system or recognizing its output as useful, without more, is insufficient. Patent applicants are not currently required to disclose use of AI in the inventive process, although the USPTO has noted that the duty of candor and good faith may require disclosure in certain circumstances. Applicants should also consider maintaining evidence of human inventorship in case inventorship is ever challenged in litigation. 

Trade secrets and AI

Given the issues surrounding the patentability of AI output, should inventors turn to trade secret protection? Trade secret protection is often used to safeguard unique intellectual property and can be obtained without application or registration. In the context of AI, trade secret protection could include protection of output, datasets, unique algorithms, and machine learning techniques.

Is AI protectable as a trade secret?

The U.S. Uniform Trade Secrets Act defines a trade secret as “a formula, pattern, compilation, program, device, method, technique, or process that: (i) derives independent economic value, actual or potential, from not being generally known to, and not being readily ascertainable by proper means by other persons who can obtain economic value from its disclosure or use, and (ii) is the subject of efforts that are reasonable under the circumstances to maintain its secrecy.”² Trade secret owners can file suit in a U.S. federal court for damages if their trade secrets have been misappropriated under the Defend Trade Secrets Act of 2016.³ In the United States, it is well established that trade secrets are property rights.⁴ 

The EU has issued a Council Directive with similar standards to the United States with regard to the definition of what constitutes a trade secret.⁵ However, the Directive generally does not regard trade secrets as property, and most EU states do not classify trade secrets as property or intellectual property.⁶

A number of issues need to be considered when applying trade secret protection to AI, including:

• Need for secrecy: Trade secrets, by definition, require secrecy. However, it can be challenging to maintain the secrecy of AI output or systems, especially in collaborative environments or open-source settings where the sharing of information and techniques is common.
• Reverse engineering: A significant drawback of trade secret protection is that it does not protect against reverse engineering or independent development, which may allow competitors to legally create their own version of an AI system’s results or the system itself. Terms of use prohibiting reverse engineering and system design limiting access to information useful for reverse engineering can help mitigate this risk in some circumstances.
• Difficult to enforce: It can be challenging to demonstrate that a trade secret has been stolen or misappropriated. For AI companies, this may require proving that a competitor had direct access to their proprietary information, which is often difficult. Outside of the United States and the EU, many jurisdictions have weak trade secret laws and/or enforcement practices.
• Employee leakage: In a tech-driven field like AI, where talent is in high demand, employees often move between companies. These employees may inadvertently or intentionally carry over knowledge or techniques that could constitute trade secrets, creating risk for companies seeking to protect their intellectual property in this way.
• Waiver of trade secret rights by disclosure to public AI: A significant risk has emerged since this guide was first published: the use of public generative AI platforms. In Trinidad v. OpenAI (N.D. Cal. Jan. 5, 2026), a federal court dismissed trade secret claims because the plaintiff had voluntarily disclosed alleged proprietary information to OpenAI while using ChatGPT to develop it. In United States v. Heppner (S.D.N.Y. Feb. 17, 2026), the court held that documents created using a public generative AI platform were not protected by attorney-client privilege because the platform’s terms of service permitted data collection and disclosure to third parties. These decisions establish that confidential information shared with public AI platforms may lose its protected status, making AI usage policies and enterprise-grade AI platforms with contractual confidentiality protections essential components of any trade secret protection strategy. 

Trade secret and best practices

While trade secret protection for AI may be challenging, much of the industry is relying on trade secret protection and adopting a “zero-trust approach.”⁷ For example, the algorithms underlying AI systems are often protected as trade secrets. Some outputs from those systems may also be kept confidential and used for the hosting companies’ commercial purposes.

Companies should implement clear policies governing employee use of generative AI tools, including prohibiting the input of trade secret information into public AI platforms and requiring the use of enterprise AI deployments with contractual confidentiality protections. They should also adopt traditional trade secret protection techniques, such as limiting access and requiring employees and independent contractors working with AI to sign confidentiality and work-for-hire agreements. Companies are also increasingly introducing methods to watermark AI-generated output to help identify the source of that content.

Notwithstanding their limitations, trade secrets have emerged as a primary intellectual property protection mechanism for AI-related assets. Unlike patents and copyrights, trade secret law does not require human inventorship or authorship, making it the most flexible vehicle for protecting AI models, training methodologies, and commercially valuable AI outputs, including when they cannot qualify for other forms of intellectual property protection. Companies should adopt a layered approach to AI intellectual property protection, combining trade secret measures with contractual provisions, including confidentiality agreements, restrictive covenants, and AI-specific usage policies.

1. ESS Entertainment 2000 Inc. v. Rock Star Videos, Inc., 547 F.3d 1095 (9th Cir. 2008).2. Uniform Trade Secrets Act (1985), Section 1.3. Defend Trade Secrets Act of 2016, Pub. L. 114–153, 130 Stat. 376 (2016).4. Ruckelshaus v. Monsanto Co., 467 U.S. 986, 1003–4 (1984).5. Directive 2016/943 of the European Parliament and of the Council of 8 June 2016 on the Protection of Undisclosed Know-how and Business Information (Trade Secrets) Against their Unlawful Acquisition, Use and Disclosure, OJ L 157, 1–18.6. Katarina Foss-Solbrekk, Three routes to protecting AI systems and their algorithms under IP law: The good, the bad and the ugly, Journal of Intellectual Property Law & Practice, 16 (3), 257, (March 2021); academic.oup.com.7. Stacy Collett, How to Protect Algorithms as Intellectual Property, CSO (July 13, 2020), csoonline.com.