Fair Use, Copyright, and Trademark Implications of Generative AI 

February 26, 2023

While many use generative AI as a fun experiment to see what it produces, others use it as a tool to complete a given task efficiently – the epitome of working smarter not harder. But just how smart is it to (arguably) rely on real life artists’ and authors’ underlying work and claim it as your own?

Generative AI is defined as artificial intelligence capable of creating text, images, or other media, using generative models. Generative AI models learn the format and pattern of its input training data and develop new outputs that have highly similar characteristics.

Some creators argue that generative AI steals their work, and they seek to debase the companies’ claim to a protective shield of fair use. OpenAI, the developer of ChatGPT, argues that these creatives misunderstand the scope of copyright protection and fail to take into account the various limitations and exceptions that provide space for innovations such as ChatGPT and other generative AI models at the forefront of the news.

In addition to the fair use piece, other copyright and trademark implications also arise for generative AI.

One must consider whether the U.S. Copyright Office will afford AI-generated content copyright protection. The U.S. Copyright Office has stated that there is no copyright protection for works created by non-humans, including machines. Therefore, AI systems cannot be considered the author of the material they produce, as their outputs are simply a culmination of human-made work. However, creative work that is the result of a collaboration between a human and a machine, which is often the case in generative AI, becomes a bit murky. The Office will consider the extent to which the human had creative control over the work’s expression and formed the traditional elements of authorship, and parse out the parts that were solely generated by the AI. Consistent with these policies, Applicants have a duty to disclose the inclusion of AI-generated content in any work submitted for registration and provide an explanation of the human authors contributions.

While much has been written about copyright issues with generative AI, the creation and use of AI-generated materials also raises a number of legal questions with respect to trademark law and trademark infringement specifically. For example, AI algorithms can be programmed to create logos or designs that closely resemble existing trademarks, potentially leading to consumer confusion or dilution of established brands. Additionally, AI-powered chatbots or virtual assistants may unknowingly infringe on trademarks when responding to customer inquiries or providing product recommendations. In the context of trademark law, there are two distinct issues for generative AI. The first is the use of another’s trademarks in the context of training materials for AI programs. Generative AI is always trained using preexisting materials. The standard test for trademark infringement is whether there is a likelihood of confusion over the source of a good or service. As such, simply using another’s trademarks to train an AI algorithm may not be capable of causing confusion where the use occurs solely internally such that there is no consumer-facing use of the trademark. However, the risk of consumer confusion is increased when trademarks are used in AI-generated outputs. If a user of an AI program takes an AI-generated mark – that incorporates or is highly similar to another’s trademark – and uses that mark in commerce, a likelihood of confusion may arise. However, again, even if an AI program is used to create a mark that is similar to an existing trademark, but that AI generated mark is never actually used in the marketplace, then trademark infringement should not be a concern because there is no real risk of consumer confusion.

Additionally, because copyright protection is not currently available for purely AI generated works, brands may consider trademark protection for AI generated marks and logos. Unlike copyright law, trademark rights are based on the extent to which a mark is used to identify the source of the goods or services. It is immaterial whether a human, an AI program, or a combination of both created or designed the mark. Currently, the USPTO has not issued any guidance indicating that AI-generated trademarks are subject to different rules than traditionally developed trademarks.

However, it is important to keep in mind that AI generated trademarks may pose an enhanced risk of infringement. As discussed, AI algorithms are trained using preexisting materials. As such, there is a risk that the program will create a mark or logo that is confusingly similar to a competitor’s existing materials. In addition to infringement concerns, purely AI generated materials may lack distinctiveness. This could result in a merely descriptive or generic refusal from the U.S. Patent and Trademark Office. Therefore, it is always best practice to have a trademark attorney review any new marks or logos for potential infringement or registrability issues.

Overall, the potential impact of generative AI on copyright and trademark law is largely speculative, as there is little case law regarding the issue. However, as AI will only continue to advance and evolve in the coming years, it is expected that courts across the country will soon be faced with these issues.

Share on LinkedIn


Madison H. McNulty



(215) 665-5562

Corinne Zucker



(215) 665-2133

Related Practices