The Army of Me and My AI

One quiet message from the U.S. government may open the AI art floodgates

While the beginning of 2025 was focused on the tectonic shift introduced by China’s Deepseek (more about that in the upcoming episode of The Hundred Year Podcast), another similarly impactful milestone emerged, its importance drowned out by a cacophony of hot takes on what the development could mean for the likes of OpenAI, Anthropic, Nvidia, and the billions in AI investment money hanging in the balance. Most in the AI space were too transfixed by the reeling stock market while simultaneously installing various iterations of Deepseek on their phones and PCs to notice that the U.S. Copyright Office made a major new decision on January 29th regarding the copyrightability of work created using AI.

A scene from Björk’s music video “All Is Full of Love” (1997)

In short, the U.S. Copyright Office has determined that AI-associated work that has been meaningfully altered or aided by the human hand can indeed be copyrighted. In contrast, work that is wholly composed of AI outputs, with no human modification, cannot be copyrighted. Nor can the prompts that generate such work be copyrighted.

Just so we’re clear, I’ve included what I think are the salient points made by the U.S. Copyright Office here:

•The use of AI tools to assist rather than stand in for human creativity does not affect the availability of copyright protection for the output.

• Copyright protects the original expression in a work created by a human author, even if the work also includes AI-generated material.

• Copyright does not extend to purely AI-generated material, or material where there is insufficient human control over the expressive elements.

• Whether human contributions to AI-generated outputs are sufficient to constitute authorship must be analyzed on a case-by-case basis.

• Based on the functioning of current generally available technology, prompts do not alone provide sufficient control.

• Human authors are entitled to copyright in their works of authorship that are perceptible in AI-generated outputs, as well as the creative selection, coordination, or arrangement of material in the outputs, or creative modifications of the outputs.

Essentially, those without a creative bone in their body, who were hoping to copyright and retain a reproducible form of ownership over their wholly AI-generated work, are out of luck. No matter how brilliantly an AI user reproduces their favorite superhero franchise using AI or generates an AI film using all of the various slick tools in their AI arsenal to create something with viral social media motion, they’re largely wasting their time and effort.

A couple of years ago, the emerging idea of a “prompt artist“ was sexy in a kind of Bruce Sterling techno-auteur as software puppeteer way. Alas, in the eyes of the U.S. Copyright Office, creators will still have to put in the scutwork of actually “making” at least some original things—a major inconvenience for those who have been planning on building a career by letting the LLMs do all of the work contained in their “artistry.“

Death of AI Art and Video Hustlers?

In contrast, this comes as incredibly welcome news for creatives who have honed their human craft to a fine point and are simply looking for a boost to their existing talents. When this fundamental protection fully seeps into the collective minds of the creative community, it will likely supercharge a new wave of human–generative AI brilliance.

However, a bit of a gray area here will challenge the artists and the U.S. Copyright Office itself. It involves this bit of language: “authorship must be analyzed on a case-by-case basis.” It seems straightforward enough, but do we really believe that the U.S. Copyright Office is currently equipped to “analyze” what is and isn’t AI-generated? At present, there are no guaranteed methods to detect AI in writing, music, visual art, or video. Solutions have been developed with varying degrees of efficacy, but AI detection in 2025 is still hit-and-miss. And no matter how well-versed one is within the realm of generative AI, we are now at the point at which even the new crop of Gen AI experts can be fooled, thus making the “eye test” from even the most experienced insiders increasingly ineffectual.

Then there’s the issue of “case-by-case” analysis. Is anyone convinced that—even if the U.S. Copyright Office had the ability to analyze and detect what is and isn’t AI—they are ready to address the coming tsunami of generative AI-assisted works that require “case-by-case” examination? I’m not. Therefore, I predict that initially, we’ll see a number of AI-assisted works copyrighted by the office as long as those works don’t somehow stand out as blatantly (a subjective conceit) AI-generated.

This triage approach could stand in while more effective AI detection methods are developed and rigorously tested. When we do have reliable AI detection, perhaps the associated Copyright Claims Board could circle back and re-examine works that have been flagged as questionable (but temporarily cleared) or perhaps challenged by a wary publisher or member of the public with an interest in certifying the human involvement in the work that buttresses its copyright.

In that case, I also predict a not insignificant number of copyrighted works from some Gen AI artists who have failed to affix their own unique human fingerprint upon the work may find their work deregistrated, delisted, and voided. Of course, this all assumes new U.S. Copyright Office processes and oversight that do not yet exist to address AI-generated work specifically. Nevertheless, as the U.S. government accelerates its familiarity with AI in general, this could evolve more rapidly than the usual glacial pace of change in the U.S. federal government.

Hidden Pioneer of AI Art Is Our Guide

But that was all looking at the dark side of the generative AI picture.

I was recently reminded of the transcendence possible when merging technology and the imperfect human touch while watching Björk’s new concert film, Cornucopia, on Apple Music Live. With a career that began in the early ‘80s and has spanned punk, art rock, and, finally, her solo endeavors as an experimental sound architect, Björk has always seemed a bit ahead of her time. She uses the most cutting-edge technology for her music and videos (via VR) to push herself to the limits of her imagination.

So it’s no surprise that one of the first AI music artists was…none other than Björk. Back in January 2020, long before most were thinking of AI creative works, and a pandemic was bearing down on the world, Bjork collaborated with Microsoft to produce a generative AI soundscape piece called “Kórsafn.” More recently, in 2024, she worked with French artist Aleph to craft an immersive AI sound installation at the Centre Pompidou in Paris called “Nature Manifesto.”

“I’ve had this discussion every time I put a drum machine on my record and people were like, ‘Oh, there’s no soul in this album,’” Bjork said following the debut of the new AI work. “The computer is not supposed to put soul into music, it’s all humans. I’ve heard a lot of soulless guitar music. We have to bring soul to things made by AI. And like all the monumental things mankind has done, we can do it.”

The ability to fluctuate between using the sounds of nature, electronic glitches, and even traditional orchestras has positioned her as a wayfinder for those who wish to follow her lead and employ technology while losing none of the elemental spirit of the human creator.

This is the exhilaratingly brighter side of generative AI creativity.

Birth of a New Kind of AI Artist

Now, fortified by the U.S. Copyright Office announcement, those artists who do command the spark of human creative genius can freely pursue new kinds of creations that are imbued with the digital steroids of generative AI. I use that term deliberately because I think there’s a meaningful analog in the world of sports.

Although performance-enhancing drugs are generally frowned upon or outright banned in some professional sports, I believe that this is a temporary dynamic. The thrill of superhuman performance on the field will eventually overtake concerns about PEDs. We’re already seeing this in the world of bodybuilding (for the uninitiated, the new film Magazine Dreams offers a sort of primer), where “natural” bodybuilders, devoid of steroids, are given their own, generally less popular, competitions. Meanwhile, the iconic images of bodybuilders typified by the likes of Arnold Schwarzenegger of old are unapologetically aided by a long list of PEDs. Contrary to what some may believe, you cannot simply take steroids and suddenly gain fabulous form and athletic ability. Even with PEDs, intense training and dedication are needed to deliver extraordinary results.

In these early days of Gen AI creativity, the same duality of effort-meets-enhancement appears to be emergent. In the ‘80s, this came through samplers in rap music. In visual art and photography, we received our first taste of digital enhancement in the ‘90s in the form of Adobe’s Photoshop and other digital tools that came after it. But generative AI is so vastly more powerful that it makes the aforementioned tools seem almost trivial by comparison. There will be those who shun generative AI tools and opt to continue to craft “artisanal” human-hand-only works, while many others produce popular works that use AI to enhance their natural artistic gifts. In many cases, the lines will blur, sometimes intentionally, as a part of the artistic expression and other times as a means of deceiving the audience.

Whether detecting generative AI and measuring its use in various works remains important to future audiences may influence the U.S. Copyright Office as it contemplates how to assign authorship to this new category of human expression. In the now, the gates have now been thrown open to those who are looking to generative AI to safely give their creative muscles extra volume, enhanced aesthetics, and that extra oomph that can spell the difference between a jejune demonstration of Gen AI’s often repetitious cruft versus genuinely sublime explorations of new corridors of the human mind’s eye.