Who Owns the Output of Generative AI Works, And Can Such Works Be Protected?

As businesses increasingly integrate generative AI tools into their workflows—from drafting marketing content to coding, product design, and visual branding—one legal question stands out: Who owns the intellectual property rights to the output produced by AI? This is more than a philosophical or academic inquiry. The answer carries serious commercial implications, especially for companies seeking to monetize, license, or enforce rights in AI-generated works. The Copyright Act of 1976 (17 U.S.C. § 101 et seq.) grants copyright protection to “original works of authorship fixed in any tangible medium of expression.” However, the statutory language presumes human authorship.

The US Copyright Office, the federal agency in charge of registering copyrights in the United States, issued guidance in March 2023 titled “U.S. Copyright Office Policy Guidance on Works Containing AI-Generated Material (March 2023), to provide clarification that works generated entirely by AI without human creative input are not eligible for copyright protection. In reviewing applications, the Office will assess the degree of human involvement in selecting, arranging, or editing AI outputs. The policy can be reviewed at https://www.copyright.gov/rulings-filings/ai-policy.pdf.

In the case of Thaler v. Perlmutter, No. 23-5233 (D.C. Cir. 2025), the U.S. Court of Appeals for the District of Columbia Circuit affirmed the lower court’s decision to uphold the denial of copyright registration for an image created entirely by an AI system. The system’s developer and owner of the image sought copyright protection but was denied because the Court held that human authorship is a bedrock requirement of copyright. Similarly, in Naruto v. Slater, 888 F.3d 418 (9^th Cir. 2018), the Court held that copyright protection should not be afforded to a photograph taken by a monkey. In both cases, the Courts held that the Copyright Act requires human authorship.

AI doesn’t just create expressive works—it can design processes, suggest product enhancements, or optimize code. Who owns those inventions? The United States Patent and Trademark Office requires human inventorship as a requisite factor in the consideration of patent applications. In another case pursued by Stephen Thaler, the appellate court held in Thaler v. Vidal, 43 F.4th 1207 (Fed. Cir. 2022), that there is a statutory requirement that individual investors must be natural persons affirming the lower court’s denial of Thaler’s attempt to gain a patent with his AI system as the inventor. Only natural persons qualify for patents under U.S. patent law.

Strategic Recommendations

Given the legal uncertainty and limited statutory protection, businesses using AI tools should adopt a layered strategy:

1. Human Involvement Is Key: Ensure that human authors provide creative input and edits to outputs before treating them as protectable works.
2. Review Terms of Use Carefully: License agreements often determine who owns what. Avoid tools that claim ownership or broad rights over your inputs and outputs.
3. Secure IP Rights by Contract: Where possible, include clauses in employment, contractor, and vendor agreements to assign any AI-assisted works to your company.
4. Keep Confidential Information Away from Public AI: Avoid entering proprietary information into public-facing AI tools that may retain or use inputs for training. This can jeopardize trade secret status.

Until Congress or the courts provide clearer guidance on AI authorship, businesses must navigate the uncertain terrain using contractual controls, internal policies, and strategic human intervention. Thorough records of the human involvement in the creative process for works that need protection should be maintained by businesses that hope to protective their intellectual property and creative assets. A proactive IP strategy can help mitigate risk, preserve value, and avoid costly legal entanglements in the future.