L. Lee: As businesses increasingly turn to generative artificial intelligence tools, or GenAI, to develop brand and marketing assets, such as names, logos, and product designs, they face novel legal questions, including who owns the copyright or trademark rights when GenAI contributes to the creation process. In BakerHostetler's third annual IP Perspectives Thought Leadership piece, BHIPP, the Intellectual Property Practice Group highlights myriad IP-related topics that are at the forefront of industry developments and current challenges and trends. I'm Leeann Lee, and you're listening to BakerHosts.
On today's episode, we discuss When Creativity Meets Code, Copyright and Trademark Troubles with AI-Generated Material. To provide us with insight into this topic, we have Eric Lee, a partner in the Intellectual Property Group at
BakerHostetler, and Terrance Roberts, an associate in the Intellectual Property Group at BakerHostetler. Both are members of the Global Trademark and Brand Protection Team. Welcome to the show, Eric and Terrance.
E. Lee: Hi, Leeann. Great to be here.
Roberts: Hi, thanks for having us.
L. Lee: Well, first of all, Eric, how is GenAI affecting clients' processes and their ability to generate and protect intellectual property?
E. Lee: Generative AI is dramatically accelerating how our clients are able to generate content and other brand assets. They are using it to help come up with names, logos, images, marketing copy, and even early product designs. That increased speed is incredibly attractive from a business perspective, but what we're seeing is a disconnect between creation and protection. So, clients often assume that because they've generated something internally using AI, that they automatically own it and that they can protect it. But in reality, as we'll talk about, the use of GenAI changes in legal analysis and requires our clients to think more carefully about authorship, ownership, and risk at the front end rather than waiting until that asset is already out in the market.
L. Lee: Great. So, let's start with copyright. Terrance, a lot of companies might assume that if they use GenAI to create an image, some text, or a product design, that they automatically own it. But that is not necessarily true, right?
Roberts: That is exactly right. And a very important consideration for creators looking to protect and control their works. Under current U.S. copyright law, copyright protection requires human authorship. So, if an image, piece of text, or other content is created by AI with no meaningful human creative input, it is not copyrightable at all. The U.S. Copyright Office has been very clear on that point, and courts have reinforced it. Just earlier this month, the U.S. Supreme Court declined to review a D.C. circuit's decision refusing copyright protection where the human copyright applicant listed an AI system as the work's author. So even though a company may possess the AI-generated output, that doesn't mean it owns enforceable copyrights. Without copyright protection, the company may have no legal ability to stop competitors from copying that content.
L. Lee: So, Terrance, it sounds like the extent of human involvement in the creative process is key here. How much human involvement is enough?
Roberts: Well, the Copyright Office has issued some guidelines, but there is no bright line rule. What matters is whether a human exercised creative control over the final expressive elements. Simply entering prompts, even dozens or hundreds of detailed prompts, is usually not enough to establish human authorship. On the other hand, if a human meaningfully selects, edits, modifies, or arranges AIgenerated material, that human contribution can be predicted.
A good example is curating and arranging multiple AI-generated images into a final composition. That selection and arrangement can be copyrightable. Practically speaking, if copyright matters, companies could treat AI output as a starting point and have humans actively shape the final result and document that involvement.
L. Lee: Okay, Terrance, let's talk about trademark rights. Is human involvement required there as well?
Roberts: So, trademark law is very different. There is no requirement that a trademark be created by a human. A company can absolutely adopt and own a name, logo, or slogan generated by AI. But the key issue in trademark law isn't authorship, it is use and consumer perception. Even if AI generates the brand, the company still has to clear it, make sure it is not confusing similar to existing trademarks, and then actually use it in commerce to build rights. AI doesn't eliminate the need for trademark clearance, and in some ways, it increases the risk because AI tools don't reliably screen for legal conflicts.
L. Lee: Great. So, let's get practical now. Eric, what should companies that are using GenAI for brand or content creation be doing right now?
E. Lee: There's a few practical steps that we're advising clients to take immediately. First, clients need to review the AI tool's terms of service. Some tools restrict commercial use or reserve rights in the outputs or allow the service to train on your outputs and your inputs. There is also data and privacy concerns to be aware of. Second, clients should implement clear internal generative AI policies so that their employees know what tools they can and cannot use and for what purposes. Third, if copyright protection is important, which it often is, clients should build in meaningful human involvement and keep records of that process to ensure copyrightability. And finally, don't skip the basics. Trademark clearance is still essential and AI-generated branding should be vetted the same way as anything created by an agency or another in-house team.
L. Lee: So overall, Eric, it sounds like GenAI can be a powerful tool, but companies need to be thoughtful both creatively and legally, right?
E. Lee: Exactly. GenAI can be very powerful, but it is not a shortcut around IP law. Companies that succeed are the ones that are going to be able to integrate GenAI into their creative workflows while still applying human judgment, legal review, and strategic oversight. That means being intentional about when AI is used, where humans step in, and how decisions are made and documented. When companies do that, AI really can be a force multiplier. It can speed up ideation, reduce costs, and unlock creativity. Without that structure, you can end up with assets that may be fast, but they're not protectable, or worse, assets that can create legal liability. The goal isn't just to create more content, it is to create content and brands that are durable, defensible, and valuable over time.
L. Lee: Well, this has been an absolutely terrific conversation. Thank you both for your insight.
E. Lee: Great. Thanks so much, Leeann.
Roberts: Yeah, thank you.
L. Lee: Thank you for this great conversation, Eric and Terrance. If you have any questions for them, their contact information is in the show notes. As always, thanks for listening to BakerHosts.
Comments heard on BakerHosts are for informational purposes and should not be construed as legal advice regarding any specific facts or circumstances. Listeners should not act upon the information provided on BakerHosts without first consulting with a lawyer directly. The opinions expressed on BakerHosts are those of participants appearing on the program and do not necessarily reflect those of the firm. For more information about our practices and experience, please visit bakerlaw.com.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.
