As artificial intelligence continues to transform industries worldwide, one of the most significant legal battles emerging in the United States concerns the use of copyrighted materials to train AI systems. More than forty copyright lawsuits have already been filed against AI developers, including major technology companies, alleging that copyrighted books, articles, music, images, and other creative works were used without authorization or compensation to train large language models and generative AI platforms.
The outcomes of these cases are expected to shape the future of artificial intelligence, intellectual property rights, and business innovation for years to come.
The Central Legal Question: Does AI Training Infringe Copyright?
According to Paul P. Cheng, Esq., Managing Partner of the Law Offices of Paul P. Cheng & Associates (PPRCLaw), former prosecutor, mediator, and veteran trial attorney, the core legal issue is straightforward:
“The fundamental question is whether the large-scale copying, storage, and processing of copyrighted works during AI training constitutes copyright infringement under existing law. While many AI companies rely heavily on the fair use doctrine as a defense, courts have not yet reached a uniform conclusion.”
The legal debate extends beyond what AI systems ultimately generate. Courts are increasingly examining whether the act of copying copyrighted content during the training process itself may violate copyright laws.
Landmark Case: Thomson Reuters v. ROSS Intelligence
One of the most significant AI-related copyright decisions to date came in February 2025, when the United States District Court for the District of Delaware issued a landmark ruling in Thomson Reuters Enterprise Centre GmbH v. ROSS Intelligence Inc.
The court found that ROSS Intelligence infringed copyrights by copying and using 2,243 Westlaw headnotes to train its AI-powered legal research platform. Importantly, the court rejected ROSS’s fair use defense.
In analyzing the traditional four-factor fair use test, the court concluded:
1. Purpose and Character of the Use
The use was commercial in nature and intended to create a competing legal research product. The court found the use insufficiently transformative.
2. Nature of the Copyrighted Work
Although Westlaw headnotes contained limited creativity, the court held that they met the minimal originality threshold required for copyright protection.
3. Amount and Substantiality Used
This factor weighed somewhat in ROSS’s favor because the copied headnotes were not directly displayed to end users.
4. Effect on the Market
The court viewed this as the most important factor. Because ROSS intended to compete with Westlaw and potentially interfere with Thomson Reuters’ licensing opportunities in the AI market, this factor strongly favored the plaintiff.
The ruling suggests that even when AI outputs do not reproduce copyrighted content verbatim, unauthorized copying during model training may still create liability, particularly when used to develop competing commercial products.
Why Fair Use May Not Be a Complete Defense
Many AI companies argue that training models on copyrighted content is analogous to search engine indexing or other transformative uses previously protected under fair use.
However, courts may distinguish AI training from those scenarios.
Paul Cheng explains:
“Fair use is not a blanket immunity for AI developers. Courts are increasingly scrutinizing whether copyrighted works are being used to advance public knowledge or simply to build commercial products that compete with the original creators. The distinction can make all the difference in litigation.”
As more AI copyright cases proceed through federal courts, businesses should not assume that fair use will automatically shield them from liability.
Can AI-Generated Content Be Copyrighted?
Another rapidly evolving legal issue is whether AI-generated content itself qualifies for copyright protection.
The U.S. Copyright Office has consistently maintained that copyright protection requires meaningful human authorship. Purely machine-generated content generally cannot be registered because it lacks human creative expression.
For businesses using AI-generated marketing materials, images, articles, software code, or branding assets, this creates significant legal uncertainty.
Paul Cheng cautions:
“Many companies mistakenly assume that because they generated content using AI, they automatically own enforceable copyrights. That assumption can be dangerous. The more human creativity, editing, judgment, and direction involved in the creation process, the stronger the argument for copyright protection.”
Businesses should carefully document human involvement in AI-assisted creative projects to help establish ownership rights.
Key Legal Risks for Businesses Using AI
Organizations integrating artificial intelligence into their operations should be aware of several emerging risks:
Copyright Infringement Claims
AI systems may generate content that is substantially similar to existing copyrighted works.
Trade Secret and Confidentiality Concerns
Sensitive company information entered into AI systems could potentially be retained, processed, or used in future model development.
Ownership and Licensing Uncertainty
The legal status of AI-generated content remains unsettled in many jurisdictions.
Compliance and Licensing Costs
If courts ultimately require licensing for copyrighted training materials, businesses may face significantly increased compliance expenses.
Best Practices for AI Compliance and Risk Management
To reduce legal exposure, businesses should consider implementing proactive AI governance measures:
- Review contracts and terms of service with AI vendors.
- Establish internal AI usage and data governance policies.
- Restrict employees from uploading confidential or proprietary information into public AI platforms.
- Maintain records of prompts, edits, revisions, and human contributions to AI-assisted works.
- Conduct legal reviews before deploying AI-generated content in commercial settings.
- Evaluate potential copyright, privacy, trade secret, and regulatory risks on an ongoing basis.
The Future of AI and Copyright Law
The law surrounding artificial intelligence is evolving rapidly. Courts, legislators, regulators, creators, and technology companies are all attempting to balance innovation with intellectual property protection.
As these landmark cases move through the judicial system, businesses that proactively address legal risks will be better positioned to avoid costly litigation and regulatory scrutiny.
Paul Cheng summarizes the issue:
“History shows that when technology advances faster than the law, uncertainty creates both opportunity and risk. Companies, creators, and innovators should not wait for every legal question to be resolved before taking action. Developing a thoughtful compliance strategy today may prevent significant legal exposure tomorrow.”
AI Copyright Litigation and Intellectual Property Attorneys in California
The litigation team at Law Offices of Paul P. Cheng & Associates (PPRCLaw) advises businesses, creators, technology companies, startups, and intellectual property owners on emerging legal issues involving:
- Artificial Intelligence (AI) Compliance
- Copyright Infringement Litigation
- Intellectual Property Protection
- Technology and Software Disputes
- Trade Secret Protection
- Business Litigation
- Content Licensing and AI Risk Management
As experienced trial attorneys, we understand that many AI-related disputes may ultimately be resolved in court. Our team is prepared to help clients assess risk, protect their rights, and aggressively advocate for their interests when litigation becomes necessary.
Contact PPRCLaw
If you believe your copyrighted works have been used for AI training without authorization, or if your business uses artificial intelligence and seeks guidance on copyright compliance, intellectual property protection, or litigation strategy, contact the experienced attorneys at Law Offices of Paul P. Cheng & Associates (PPRCLaw) for a confidential consultation.
Disclaimer: This article is for informational purposes only and does not constitute legal advice. Every case is unique. Consult qualified legal counsel regarding your specific situation.