The United States Copyright Office (USCO) receives a legion of copyright requests annually. Since generative artificial intelligence entered the scene, the office has seen increasing requests to copyright AI-generated or AI-assisted creative works.

A recent report concludes that copyrightability depends on the extent of AI or human involvement in the creative process and finished product. Here’s everything you need to know about AI-generated outputs and copyright in the United States.

What Does the USCO Copyright Report Actually Say?

The USCO launched an initiative back in 2023 specifically to examine copyright law and policy issues introduced by artificial intelligence. In July of 2024, they released Part 1: Digital Replicas with Part 2: Copyrightability to follow in January 2025.

While the focus here is on copyright, here’s a brief overview of Part 1: Digital Replicas, should you find it interesting:

Part 1: Digital Replicas Overview

The report defines digital replicas as AI-generated representations that realistically mimic an individual’s likeness or voice, such as for entertainment or advertising purposes. While existing laws offer some protection against the unauthorized use of an individual’s likeness, they vary by state and may not comprehensively address the challenges posed by digital replicas.

Part 1 of the report outlines recommendations for federal legislation to address the unauthorized creation and distribution of digital replicas, such as:

  • Establish clear definitions: Clearly define what constitutes a digital replica.
  • Set consent requirements: Mandate obtaining explicit consent from individuals before creating or distributing their digital replicas.
  • Provide remedies: Offer legal recourse for individuals whose likenesses have been used without authorization.

There’s a clear emphasis on balancing technological innovation with individual rights — which is a theme that partly carries over to the Copyrightability report.

Part 2: Copyrightability Overview

In Part 2 of the report, the USCO examines the application of existing copyright principles to works created using generative AI technologies. In general, it reaffirms that copyright protection is reserved for works that are the result of human creativity. Purely machine-generated content, without significant human involvement, does not qualify for copyright protection.

That doesn’t mean you can’t copyright AI-assisted work, though. When a human uses AI as a tool to assist in their creation, the resulting work may be eligible for copyright protection — so long as the human contribution is substantial and evident in the finished product.

People are using a Jackson Pollock analogy to clarify what this means:

Pollock, famous for his “drip technique,” didn’t decide how the paint landed on the canvas, but he was undoubtedly in control the whole time, and his finished pieces were unquestionably his. The USCO says, “The issue is the degree of human control, rather than the predictability of the outcome.”

But it gets even more granular. Here are a few key stipulations that the report discusses:

  • Prompts: Merely providing prompts to an AI system, regardless of their complexity, does not constitute authorship of the generated output and therefore has no solid grounds for copyrightability.
  • Modification: Human creators who modify AI-generated content in a way that reflects their own creative expression may be able to obtain copyright protection for the modified work.
  • Flexibility: The report acknowledges the evolving nature of AI, ensuring that copyright laws are flexible enough to address issues arising from the use of AI in creative processes — particularly as the tech becomes more capable of expressive output.

Succinctly, this means that if an artist’s personal touch is evident in the final product, it may qualify for copyright protection — at least parts of it.

If you’re interested in a visual example, see page 23 of the report, which shows a case where only parts of a submitted work were eligible for copyright protection.

“Where a human inputs their own copyrightable work and that work is perceptible in the [AI] output, they will be the author of at least that portion of the output. Their own creative expression will be protected by copyright, with a scope analogous to that in a derivative work,” writes the USCO.

How Do People Feel About AI-Assisted Creations Enjoying Copyright Protections?

While it’s difficult to determine right now how people feel about copyright as it relates to AI-assisted works, it’s pretty safe to say that many artists aren’t happy with AI art in general, no matter the capacity of human involvement.

Christie’s, an age-old auction house in the UK, recently announced its first AI-generated art auction. Artists aren’t happy, and an open letter to the organization expressing concern and calling to cancel the event has garnered over 5,000 signatures at the time of writing this blog post.

Like anything else, AI and copyrightability will draw strong debates from all viewpoints. That said, I believe the USCO has the right idea: Unless you can prove significant human involvement or the creator’s artistic vision is clearly expressed in the finished work, no copyright.

How USCO’s Approach to AI Copyright Could Impact Brands and Marketers

While USCOs approach to AI and copyright is more or less a win, in my opinion, it does introduce some technicalities that brands and marketers currently using — or that plan to use — AI in their creative processes should be aware of:

  • AI-generated content isn’t automatically protected: If a brand creates marketing materials using generative AI (e.g., AI-generated images, videos or text), those materials may not be protected by copyright unless significant human creativity is involved — and evident.
  • There’s a risk of imitation: Since AI-generated content without human authorship isn’t eligible for copyright, competitors can legally use, modify or copy similar AI-generated works.

So, if you’re looking to protect your brand properties on all fronts, human authorship and sustained involvement in creative processes are absolutely necessary. This could include:

  • Customizing AI-generated visuals with human-designed elements.
  • Editing AI-generated text to reflect unique brand messaging.
  • Applying creative judgment in selecting and refining AI outputs.

Here are a couple of context-specific examples to help paint a clearer picture:

Brand Identity & AI-Generated Logos

If a brand uses AI to create its logo, slogans or other core brand assets, it might not be able to claim copyright protection. This means it could be harder to legally prevent others from using similar branding.

Content Licensing & AI in Advertising

Marketers who use AI-generated content should be cautious when licensing or selling content. If it lacks copyright protection, they may not have exclusive rights over the work, which can cause problems for their brand and even the client. Advertisers should clearly define AI-related terms in contracts to avoid disputes over ownership and usage rights.

The Bottom Line for Brands & Marketers

AI-generated content alone isn’t enough to earn copyright protections; human creativity continues to play a key role in the copyrightability process:

“Copyright law has long adapted to new technology and can enable case-by-case determinations as to whether AI-generated outputs reflect sufficient human contribution to warrant copyright protection. As described above, in many circumstances these outputs will be copyrightable in whole or in part—where AI is used as a tool, and where a human has been able to determine the expressive elements they contain. Prompts alone, however, at this stage are unlikely to satisfy those requirements,” writes the USCO.

For marketers and brands interested in originality and protecting their digital and intellectual properties, human creativity prevails once again.

Note: This article was originally published on contentmarketing.ai.