These questions underscore a central legal dilemma: current intellectual property laws are struggling to keep pace with the unique challenges posed by AI-generated content. Traditional frameworks and precedent for copyright, patent, and trademark protection were conceived with human creators in mind, leaving uncertainty about how to protect, license, and enforce rights in works generated by, or with the assistance of, AI. Nixon Peabody serves as a forward-thinking partner for businesses exploring AI and intellectual property opportunities.
How AI is challenging traditional IP frameworks
Artificial intelligence challenges the traditional intellectual property (IP) framework by blurring the distinction between human and machine authorship and raising questions about who—or what—should be recognized as a legal author or inventor. This ambiguity creates significant IP issues with AI due to gaps in existing IP law, as most statutes do not recognize non-human creators, leaving courts and policymakers grappling with how to allocate rights and responsibilities for AI-generated outputs.
Existing copyright, patent, and trademark laws often fail to address the nuances of AI-generated content. For example, the US Copyright Act of 1976 requires a human author, while patent law demands a human inventor. High-profile developments, such as the US Copyright Office’s refusal to register works created solely by AI and the international debate over whether an AI system can be listed as an inventor on a patent application (as seen in the DABUS patent cases), highlight this legal grey area.
AI-generated content and ownership issues
Who owns content created by AI—the user who prompts the AI, the developer who built the system, or perhaps no one at all?
AI-generated works often lack a clear human creator, blurring traditional notions of authorship and ownership. As noted above, content generated solely by AI is ineligible for copyright. However, leaving such works in the public domain raises concerns about control, attribution, and commercial use.
Industry observers and legal professionals have discussed ;potential workarounds to establish AI-generated content intellectual property rights. One method these stakeholders have considered is to establish a threshold of human input—if a user’s creative choices are sufficiently significant, the resulting work might be eligible for copyright protection, with the user as the author. Another method is to recognize co-authorship, where both the user and the AI developer share rights in the work, especially if the developer’s role in designing the AI is deemed integral to the creative process. With these AI intellectual property issues still so novel, there is no precedent in place, and the law therefore remains unsettled.
AI and intellectual property law: Current legal landscape
Courts, lawmakers, and intellectual property lawyers worldwide are actively grappling with the challenges that artificial intelligence poses to traditional IP frameworks. The DC Circuit has held that works generated solely by AI, without significant human involvement, are not eligible for copyright protection, and the US Patent and Trademark Office (USPTO) has rejected patent applications listing AI as the sole inventor, maintaining that inventorship must be attributed to a natural person. Global courts are weighing these issues as well:
- In November 2023, the Beijing Internet Court in China recognized copyright protection for an AI-generated image, provided there was demonstrable human intellectual effort involved.
- The European Union’s AI Act introduces transparency requirements for AI training data and aims to balance copyright protection with innovation.
Clarifying the legal landscape
The US Copyright Office has issued guidance requiring 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 analysis. The USPTO is similarly engaged in ongoing reviews and public comment periods to address questions involving AI-assisted inventions. Internationally, organizations like the World Intellectual Property Organization (WIPO) are studying the benefits of AI for intellectual property, the economic and legal implications of AI use cases in intellectual property systems, and encouraging harmonization of standards. These efforts will likely be pivotal in shaping balanced and adaptive IP protocols.
Generative AI tools can introduce risk
Generative AI tools rely on vast datasets scraped from the internet, much of which includes copyrighted material. While diverse, high-quality data is needed to train models capable of producing human-like outputs, obtaining licensing for every piece of data is logistically challenging and expensive. As a result, many developers choose to use publicly available data without a license under the assumption that it falls under fair use or similar doctrines.
This assumption has sparked lawsuits against AI developers alleging that the unlicensed use of copyrighted works in training datasets constitutes infringement. Courts have explored fair use defenses, and the US Copyright Office has issued reports setting out a fair use analysis for AI training. Courts’ analyses of cases of first impression could have far-reaching implications for the AI industry, potentially requiring companies to overhaul their data collection practices, pay licensing fees, or limit the capabilities of their models.
Risk mitigation strategies for businesses using AI
In the US, Congress has introduced the Generative AI Copyright Disclosure Act of 2024, which would require developers of generative AI models to disclose the datasets used for training to increase transparency and give copyright owners more control.
To minimize exposure when deploying generative AI tools, businesses and creators should assess the terms of service and licensing agreements associated with any AI platform they intend to use and minimize intellectual property risks associated with the data used to train the platform. Businesses should also implement robust internal policies that require human oversight in all content creation workflows, ensuring that AI-generated outputs are reviewed for compliance with legal, ethical, and brand standards before publication.
Involving intellectual property lawyers early in the development of processes incorporating AI can help identify potential risks related to copyright issues. Working with counsel to establish best practices and clear review processes can help safeguard businesses against potential disputes involving generative AI.
The future of AI and IP: What’s next?
Looking ahead, legislation clarifying the intersection of AI with intellectual property rights seems inevitable as societies grapple with the complexities introduced by machine-generated content. We may see novel categories of IP protection emerge, specifically for works created by autonomous systems, and clearer guidelines for assigning liability when AI systems infringe upon existing rights. As creators, businesses, and legal professionals capitalize on the transformative potential of AI, it is crucial to engage with an artificial intelligence IP attorney to understand the risks associated with rapidly changing technology and to navigate regulations that have yet to catch up. Strong legal guidance can help businesses and creators minimize exposure to liability while seizing the opportunity to break new ground in this dynamic landscape. Connect with Nixon Peabody's IP team for strategic insight and legal clarity.
FAQs
Can I register a copyright for content created using generative AI?
Fully AI-generated content is ineligible for copyright registration in the US, as human authorship is required. The US Copyright Office’s March 2023 guidance states that “If a work’s traditional elements of authorship were produced by a machine, the work lacks human authorship and the Office will not register it.”
Is it legal to train AI models on copyrighted material?
Use of copyrighted material for training AI models often relies on the doctrine of “fair use,” a principle that permits limited use of copyrighted material without permission from the copyright holder, often for purposes such as journalism, teaching, or research. Courts consider several factors to determine fair use, including the purpose of the use (e.g., commercial vs. non-commercial); the nature of the copyrighted work; the amount and substantiality of the content used; and market impact.
There is ongoing debate about whether using copyrighted material to train AI models is “transformative” and could qualify as fair use, but ongoing judicial interpretation will be needed to determine whether such use is permissible.
Can AI-generated inventions be patented?
USPTO guidance clarifies & that AI alone cannot be named as an inventor, and AI-generated inventions can be patented only if a human makes a “significant contribution” to the invention. However, patent protection is available for AI-assisted inventions when human input meets the legal standard for inventorship. To understand these standards and how they can protect your AI-assisted invention, reach out to Nixon Peabody’s intellectual property lawyers.
How are global IP laws responding to AI-generated content?
Global IP laws are in flux as courts and lawmakers address challenges posed by AI-generated content. In the US, works created solely by AI, without meaningful human input, are not eligible for copyright or patent protection, and the Generative AI Copyright Disclosure Act of 2024 aims to increase transparency around AI training data. The EU’s AI Act also emphasizes transparency while balancing the goals of innovation and robust intellectual property rights. China recognizes AI-generated works when there is clear human intellectual effort. International bodies like WIPO are aiming to synchronize standards and study the broader implications of AI for IP frameworks. Overall, there is a global trend toward greater legal clarity in how to define and protect AI-generated works.