Copyright Protection in the Age of AI: An Interview with Meaghan Kent

Meaghan Kent (Morgan, Lewis & Bockius, USA)

The proliferation of generative artificial intelligence (AI) has opened up many important discussions concerning how the use of tools such as ChatGPT and the products they create intersect with various intellectual property (IP) rights, including copyright protection. As jurisdictions from around the world develop legislation and regulation efforts surrounding AI, there is no uniform approach, but trends are beginning to emerge.

Meaghan Kent (Morgan, Lewis & Bockius, USA) will moderate a session titled A Comparative Review of the Interplay Between Copyright and AI in Various Jurisdictions at the upcoming 2024 Leadership Meeting taking place next month in New Orleans, Louisiana, USA. The session will tap into expertise from several jurisdictions to provide an overview of litigation, policy, legislative, and regulatory issues regarding AI and copyright protection—including the use of copyrighted works to train models, the question of whether generative AI outputs constitute copyright infringement, and the question of how much human authorship is required for copyrightability of AI-assisted outputs.

In this interview with the INTA Bulletin, Ms. Kent sheds some light on the emerging trends in regulation across the globe, delves into both sides of the argument over whether copyrighted materials should be made available for training generative AI programs, and encourages copyright holders to be nimble and consider their options when it comes to AI.

You will be leading a session at the 2024 Leadership Meeting entitled “A Comparative Review of the Interplay Between Copyright and AI in Various Jurisdictions.” Please provide a brief overview of what registrants can expect from this discussion.
We will have panelists from the worlds of academia, government, and private practice, including from Asia, Europe, South America, and the U.S., providing a wide range of experience and perspectives on this very complicated and developing area of the law.

The panelists will discuss regulatory developments as well as pending cases that cover the important questions of potential infringement concerns and copyrightability concerns—including whether AI training using copyrighted content infringes the copyright in that content, or whether there may be carve-outs in the regulatory frameworks or by case law under the fair use doctrine. We’ll also cover whether generative AI outputs from tools that were trained using copyrighted content can infringe that underlying content and whether content created by generative AI can be protected by copyright and in what circumstances.

ChatGPT became publicly available AI at the end of 2022. Though AI has been around for many years, broad access to generative AI marked a paradigm shift in the AI space with far-reaching implications, including for IP owners and the IP community. What are some of the most recent developments in AI regulation in terms of copyright protection?
In the session, we’ll discuss regulations and pending regulations being implemented in various jurisdictions. This will include the EU AI Act, which has garnered significant attention and is expected to have far-reaching impact given the language that allows for broad application, including the obligations and exceptions articulated in the Act. We will also discuss updates within the U.S., including the AI Executive Order, the Copyright Office Studies, potential Compendium updates, the Senate AI Policy Roadmap, and introduced legislation such as the Generative AI Copyright Disclosure Act.

Another recent development we hope to discuss is Brazil’s AI legal framework and others that will impact where companies may choose to train their AI models. Singapore has carved out a Computational Data Analysis exception that is attracting some companies, and other similar regulations are being introduced in places like Japan.

AI at the 2024 Leadership Meeting

Registrants at the 2024 Leadership Meeting looking to learn more about AI are encouraged to attend the following sessions:

Policy Perspectives: Elections and the Impact on the Global Intellectual Property Landscape
Wednesday, November 13, 10:00 am–11:00 am
Room: Empire B

Join us as INTA’s Chief Policy Officer leads a fireside chat with INTA’s Chief Representatives from various jurisdictions to discuss post-election implications on intellectual property. This year, the world witnessed an unprecedented number of people going to the polls. Gain valuable insights and perspectives on how potential political shifts may affect international IP treaties, harmonization efforts, the regulatory race around AI, and cross-border enforcement, and how this could impact your organization.

A Comparative Review of the Interplay Between Copyright and AI in Various Jurisdictions
Wednesday, November 13, 2:30 pm–3:30 pm
Room: Empire B

Panelists from the worlds of government, academia, and private practice will discuss regulatory developments as well as pending cases that cover the important questions of potential infringement concerns and copyrightability concerns This session will tap into expertise from Europe, South America, and the United States to provide an overview of litigation, policy, legislative, and regulatory issues regarding AI and copyright protection—including the use of copyrighted works to train models, the question of whether generative AI outputs constitute copyright infringement, and the question of how much human authorship is required for copyrightability of AI-assisted outputs.

Navigating the Complexities of Trade Secret Protection in the Modern Era
Wednesday, November 13, 3:45 pm–4:45 pm
Room: Empire B

Protecting trade secrets has become increasingly complex in today’s rapidly evolving digital landscape. This session will delve into the unique challenges organizations face in safeguarding their proprietary information. Our discussion will cover the following critical areas:

Managing access to sensitive information.
Balancing secrecy with accessibility.
Regulatory aspects of using trade secrets in AI.

Join us for an insightful session that will equip you with the knowledge and strategies to effectively protect your organization’s trade secrets in the modern era.

Fighting Online Scams in the AI Era
Thursday, November 14, 10:00 am–11:00 am
Room: Empire B

Scams are a significant threat in the digital era, with online scams causing substantial financial losses and damaging brands’ reputations across various industries. As these scams evolve, they are increasingly leveraging AI to create more sophisticated and convincing fraudulent schemes. This session will delve into the nature of online scams, their impact on businesses, and the latest trends shaping this issue. We will also explore AI as both a threat and an opportunity in the fight against online scams and discuss effective strategies and best practices for scam prevention. We will also address whether the challenge is more in the wheelhouse of the legal team or the security/IT team of an organization.

Ethical and Practical Considerations for Incorporating AI Technologies in Your Trademark Practice
Friday, November 15, 12:00 pm–1:00 pm
Room: Empire B

Join our panel of AI experts and legal ethics gurus to learn about the practical use, potential pitfalls, and ethical concerns associated with using (or not using) AI-based technology in your trademark practice. Using specific situations to describe where you may, could, or should use AI-based technology, our panelists (along with ChatGPT) will guide you through the questions to ask yourself—and others—before integrating AI tools into your legal practice. Learn practical ways to evaluate and analyze new technologies and how to responsibly integrate AI technology into your professional practice.

Could you broadly speak to a few of the different approaches currently being taken by jurisdictions regarding copyrights and AI?
Several jurisdictions have or are developing frameworks and exceptions for AI training, including Brazil, the EU, Japan, and as previously mentioned, Singapore. The U.S., though, is currently allowing the courts to address some of these questions and the Senate has specifically said in its AI Policy Roadmap that while it needs to “[c]onsider federal policy issues related to the data sets used by AI developers to train their models, including data sets that might contain sensitive personal data or are protected by copyright, and evaluate whether there is a need for transparency requirements,” that it is first going to review and await the results of the “existing and forthcoming reports from the U.S. Copyright Office and the U.S. Patent and Trademark Office on how AI impacts copyright and intellectual property law, and take action as deemed appropriate to ensure the U.S. continues to lead the world on this front.”

One of the major questions brand owners are currently asking is whether or not copyrighted materials should be made available to train generative AI models. What are the major arguments both for and against this practice?
Content creators argue that as the copyright owners, they have the right to control the copying, distribution, and creation of derivative works and that the use of their content for training without their permission infringes on those rights. The companies using the content for training have for the most part been relying on arguments of fair use, including that the use is transformative. Underlying these arguments are practical and arguably existential concerns about how this will impact creatives and how the case law and regulations can continue to encourage creativity and innovation in both the arts and technology.

Other factors that are overlapping and perhaps running parallel to this are increasing issues related to rights of publicity, digital replicas, and so-called deep fakes and the need for AI-generated works to be conspicuously identified.

What advice would you offer copyright holders right now in regard to AI?
Attend our panel as a first step! The developing laws and regulations surrounding copyright and AI require creators and innovators to be nimble and consider their various options in various jurisdictions. This session will provide an overview of these options.

Learn more and register for the 2024 Leadership Meeting.

Although every effort has been made to verify the accuracy of this article, readers are urged to check independently on matters of specific concern or interest.