What I’m Working On: Professor Kristelia García, Intellectual Property Expert
October 10, 2024
Having practiced entertainment law for nearly a decade before joining academia, Professor Kristelia García still finds scholarly inspiration in the music industry — especially when it comes to cutting-edge issues at the intersection of copyright, law and economics.
“I follow a lot of industry blogs and pay attention to the deals taking place, particularly the ones that make me think: ‘Why is that happening?’” she says. “Where the law predicts one behavior but what we see playing out in the real world is different — that’s when I know I’ve got a potential research project.”
Below, García discusses testifying to lawmakers on Capitol Hill about artificial intelligence (AI) and intellectual property rights, how the misapplication of copyright law harms emerging artists and the Taylor Swift deal that inspired her first scholarly work.
You recently testified at a hearing about intellectual property and AI held by the U.S. House of Representatives Judiciary Subcommittee on Courts, Intellectual Property, and the Internet. Why is there concern about granting copyright to works that are wholly or partially AI-generated?
Right now, there’s a lot of buzz about AI, but it’s hard to know how much of it is genuine concern based in fact, and how much of it is just worry about the unknown. Generally, if someone writes a book or makes a film, they get a copyright on that work. Copyright effectively sets up a “mini-monopoly” for that product by giving the copyright owner the exclusive right to sell or license that work. In that way, the law affords creators a way to make a living, which (hopefully) incentivizes them to create.
Along comes generative AI, which can create entirely new works from the raw data it is trained on. This raises the question: Can you copyright wholly AI-generated works? It also raises the questions: What makes something creative? Is creativity strictly human? Can AI be creative? We give copyright protection to all sorts of things that, from a layman’s perspective, are not that creative. The intersection of generative AI and copyright is really interesting because it tugs at all of these aspects of copyright doctrine and challenges the way we think about them.
What view do you and other legal scholars take?
There are (at least) two different takes. My view is that we shouldn’t grant copyright protection to wholly AI-generated works for a variety of reasons that I discussed at the hearing and in my written testimony, including the fact that AI can’t be incentivized to create and doesn’t meet the statutory definition of an “author” under the Copyright Act. When it comes to partially AI-generated works, maybe we should grant copyright under certain circumstances, such as when humans are sufficiently involved in prompting the AI to create. But within that camp, there are different opinions when it comes to questions like: What is being created? And how creative is it?
What myself and other speakers tried to highlight at the hearing was that there’s not a lot to be gained by giving protection to wholly AI-generated works at this time. On one hand, we don’t want to discourage artists from using these tools and doing cool things with them. But we want to balance that with how much these works deserve, and even need, protection.
On the topic of copyright, your article “The Emperor’s New Copyright” was published last fall in The Boston University Law Review. What does that paper explore?
That piece focuses on cases in which people misuse or misapply copyright law, either intentionally or unintentionally. For example, within the film industry, there’s something commonly called “clearance culture,” where production companies and studios require that filmmakers “clear,” or get permission, to use all sorts of material — such as film samples or music clips — that may not actually require licensing in order to avoid claims of copyright infringement. To do so, filmmakers have to use copyright clearance companies that clear all of this stuff unnecessarily. This really impacts small filmmakers who would like to be able to create work but can’t afford a massive clearance process. As a result, clearance culture can reduce innovation and creation, because less-resourced filmmakers can’t get anything done.
You came to academia having practiced entertainment law in the music industry. How are legal scholarship, policymaking and practice linked for you?
I’ve always been very policy-oriented in my work and writing. A large part of what attracted me to Georgetown Law was the chance to be closer to the action in terms of policy and lawmaking in Washington, D.C. Without a doubt, the work that I did in practice heavily influences what I do as a scholar. In fact, the very first piece I wrote when I was making the transition to academia drew on a matter I learned about while at Universal Music Group, where I worked on the digital licensing for Taylor Swift’s “Red” album.
One of the things I explored in that paper is how influential artists like Swift can negotiate certain rights for themselves that don’t actually exist under the statute. For instance, we worked on negotiating what’s called a “terrestrial performance right” for her, which is a fancy way of saying that she would get paid when her songs were played on certain radio stations. As the law stands, when a song is played on the radio, the songwriter — not the recording artist, who is often a different person — receives royalties.
In negotiating that deal, Swift’s label circumvented a little known and often-overlooked section of the Copyright Act — which says that a portion of digital streaming royalties go to the session musicians, who often rely on those payments — by negotiating a lower digital rate in exchange for the terrestrial performance right, among other things. This type of deal highlights the risk that other artists might also circumvent the statute and cut these musicians out of the picture. Since then, I haven’t used specific deals in the same way, but I still regularly draw on real-world examples in my work.
What are you working on next?
I’m currently editing an article about selective enforcement, or how copyright owners decide whether or not to sue for copyright infringement — because sometimes they do, and sometimes they don’t. This project digs into the question of why this is happening and what the implications are.
I’m also beginning work on another project that examines the intersection of antitrust and copyright, which I’m tentatively calling “Refusal to License.” Anticompetitive concerns can arise when claims of copyright infringement are used as a shield for anti-competitive practices, such as when Spotify bought the podcasting company Gimlet and then pulled Gimlet’s content from other podcasting platforms. I’m collaborating on that project with a former copyright student of mine, and it’s been really fun.