How Have New Developments in AI Shaped Patent and Intellectual Property Law?

The evolutions and stark increases in artificial intelligence over the past couple of decades arguably represent the largest confrontation humanity has faced in such a short span of time. Seeping into every sector, discipline, and structure, the rise of AI and its subsequent impacts on innovation has been accompanied by widespread doubt over the threats it poses to human creativity. One large manifestation of such fears is the new role AI plays in intellectual property (IP) and trademark jurisdiction. 

The concept of owning an idea has deep historical roots, with sources citing it back to the Enlightenment, 13th century, and even Ancient Greece. At its core, intellectual property arises from the belief that humans are creators, and thus owners of thought. Based and designed on the preservation of human authorship and creativity, intellectual property law rewards individual human inventors and artists. The rise of generative AI blurs this boundary, as courts are now grappling with questions on consolidating and applying traditional IP frameworks to works generated or assisted by AI. 

As such, emerging case law suggests two competing considerations courts must now reconcile. First, AI as a creator: distinguishing between human and machine output in applying IP authorship to the creation of works. Second, AI as a threat to creativity: holding generative AI firms accountable for unlawful use of training data and reproducing copyrighted work. 

AI as a Creator

The first emerging consideration is how to define the boundary between human and machine authorship: what can be patented, who owns the rights, and what counts as a legal author or inventor: the prompter? The machine itself? Its creator? When artificial intelligence is used as a crutch in the creation of a work, how can courts make tangible a collaboration between human and algorithm? 

Existing legislation only considers works made by humans and fails to address the nuances of AI-generated content: a human author is required in the U.S. Copyright Act of 1976 and a human inventor is required in existing patent law. 

Courts are careening toward a definition, albeit a nascent one: the USPTO guidance, for example, specifies that AI alone is not enough to be recognized as an author, and that the significance of human contribution must rise above a certain threshold to warrant protection under IP law. The level of significance of human input required to meet the legal standard for inventorship remains unclear and is left to the best judgment of individual courts. 

One can only expect emerging case law and precedent to draw this distinction more clearly. Already, the U.S. Copyright Office requires applicants to disclose any AI-generated content in works submitted for registration, and continues to refine its policies through public consultations and case-by-case decisions.

AI as a Threat to Creativity 

The second trend at the forefront of recent case law, perhaps a more nuanced one, involves the issues courts confront regarding the creative threats AI poses to existing works. In the past couple years and months, the judiciary has been flooded with AI infringement cases, including Reddit v. Perplexity and New York times v Open AI. Of them, OpenAI and ChatGPT alone have already been subject to 20 ongoing lawsuits. 

Reddit chief legal officer Ben Lee referred to this trend in a statement, "AI companies are locked in an arms race for quality human content - and that pressure has fueled an industrial-scale 'data laundering' economy." 

Most of the cases revolve around copyright infringement by way of firms scraping data unlawfully to train their artificial intelligences. However, the list of claims doesn’t stop there. Intertwining a wide range of IP-related violations, these emerging lawsuits, through both similarities and particularities, shed light on the broad spectrum of legal implications and damages companies can assert under IP law and thus the broad spectrum of ways in which AI poses threats to claimants.

Ziff Davis v. OpenAI, for example, layers a trademark dilution claim onto its copyright infringement allegations, asserting that OpenAI’s use of its content harms the reputation of its trusted brands: “…by misattributing content by Ziff Davis to other publishers or content by other publishers to Ziff Davis, OpenAI is undermining the public’s perception of Ziff Davis.” (Ziff Davis v. OpenAI, Section 197).

Implications

These legal developments remain in their early stages, with most lawsuits still at the outset. Yet the cases and questions courts are beginning to confront carry significant implications for the future of AI, the future of IP law, the role of human creativity within these evolving frameworks, and, more broadly, how we as a society value human creation. 

The precedent that emerges from these cases will help clarify both the infringement capabilities of AI and the methods courts can use to identify unlawful data scraping. While Ziff Davis v. OpenAI is one of the few cases providing explicit examples of AI-generated content copying the plaintiff’s speech, most suits lack such concrete evidence, leaving courts with the difficult task of determining infringement without direct textual parallels.

Another emerging precedent concerns how firms can avoid these claims, both in building AI systems and in outsourcing them. Internationally, organizations such as WIPO are studying how AI interacts with IP systems and are pushing for greater harmonization of standards. Collectively, these efforts will shape how companies are expected to navigate compliance and maintain transparent, defensible practices. 

In the U.S., the proposed Generative AI Copyright Disclosure Act of 2024 would require developers to disclose the datasets used to train their models, increasing transparency and giving copyright owners greater oversight. For companies using AI, this likely means reading terms of service and licensing agreements carefully, implementing internal policies that ensure human oversight, and involving IP counsel early in the development of any AI-driven processes.

Taken together, these emerging legal battles signal more than a shift in doctrine: they force us to confront how we value human imagination, how we distribute credit in an algorithmic era, and how we safeguard creative ecosystems. As courts, companies, and policymakers adapt, the choices they make will shape not only the contours of IP law, but the broader social understanding of what it means for humans to create in a world where machines can, too.