Recognizing AI-Assisted Art: The Copyright Office is Using the Wrong Legal Standard

“The standard is whether there is a ‘modicum of creativity,’ not whether Kashtanova could ‘predict what Midjourney [would] create ahead of time.’ In other words, the Office is incorrectly focusing on the output of the tool rather than the input from the human.”

op-edThe author is representing Kashtanova in the case discussed below, along with Max Sills of Open Advisory Services.

The U.S. Copyright Office (USCO) released its decision this past week in Kristina Kashtanova’s case about the comic book, Zarya of the Dawn. Kashtanova will keep the copyright registration, but it will be limited to the text and the whole work as a compilation.

In one sense this is a success, as the Office was previously threatening to revoke the copyright altogether. But the Office limited the registration and specifically excluded the individual images created by Kashtanova from the copyrighted material. This is a setback for all the artists that would like to use artificial intelligence (AI) tools as part of their creative process.

Taken in the context of the other AI-driven cases the Office is fielding, this is a surprising result. In the USCO’s recent filing in the Stephen Thaler case, the Office said that it was “preparing registration guidance for works generated by using AI.” If the Office is preparing guidance to allow registration of AI-assisted works, that strongly suggests that the USCO believes there is some threshold of human involvement that is sufficient to allow registration.

The Office recognizes that Kashtanova had input into the images that were created for Zarya of the Dawn, but it argues that the human input had no substantial effect on how the images came out: “the process is not controlled by the user because it is not possible to predict what Midjourney will create ahead of time” (p. 8) or “Rather than a tool that Ms. Kashtanova controlled and guided to reach her desired image, Midjourney generates images in an unpredictable way.” (p. 9) [Editor’s note: The USCO’s letter incorrectly refers to Kashtanova using she/her pronouns; Kashtanova’s pronouns are they/them].

A ‘Modicum’ of Input is All it Takes

This isn’t the right legal standard. The standard is whether there is a “modicum of creativity,” not whether Kashtanova could “predict what Midjourney [would] create ahead of time.” In other words, the Office is incorrectly focusing on the output of the tool rather than the input from the human.

Jackson Pollack famously couldn’t predict how the paint he used would drip onto the canvas. Pollack designed his paintings but he used a process involving random dripping and flicking of paint to make his art. In music, each performance of John Cage’s 4’33” is entirely defined by the random sounds that are made by the audience and the world around the stage.

In photography, the closest analogous art, there are famous photographs that captured animals, people, or humorous situations entirely by mistake. Nevertheless, the output is still copyrightable, because a human had at least a “modicum” of input into the shot.

Even more concerning to AI artists are the factual misunderstandings that drive this decision. The Office appears to think that the defining characteristic of AI tools is that the output is “random” or “unpredictable.” As stated by the Office, “The fact that Midjourney’s specific output cannot be predicted by users makes Midjourney different for copyright purposes than other tools used by artists.” (p.9)

The exact output may not be predictable, but experimentation will show that when you type “cute purple dinosaur” into Midjourney, you get back images of a cute purple dinosaur, not a motorcycle or a cloud. Further, the more inputs given by the artist, the more control is exerted over the output. The artist can design the output to have a specific subject, lighting, content, layout, and feel. It should not matter that the subject, lighting, content, and layout are generated instead of captured.

It is still early for AI-assisted art and the USCO is engaging in a number of outreach events to get feedback. But ultimately, AI-assisted art will need to be held to the same standards as photographs, where even prosaic photos have enough human input to be copyrightable.

Image Source: Deposit Photos
Image ID: 31248541
Author: tashatuvango

Share

Warning & Disclaimer: The pages, articles and comments on IPWatchdog.com do not constitute legal advice, nor do they create any attorney-client relationship. The articles published express the personal opinion and views of the author as of the time of publication and should not be attributed to the author’s employer, clients or the sponsors of IPWatchdog.com.

Join the Discussion

7 comments so far.

  • [Avatar for Alan Harrison]
    Alan Harrison
    February 27, 2023 07:05 am

    Can copyright be established in art produced by https://en.wikipedia.org/wiki/Spirograph?

  • [Avatar for AIA]
    AIA
    February 26, 2023 03:24 pm

    As someone who has experience with AI art, the random part hit the nail right on the head. The process feel more like doing a search on Google images, there is no real control of the output, you just settle with the option that more or less you had in mind.

  • [Avatar for Josh Masur]
    Josh Masur
    February 26, 2023 12:41 pm

    Burrow-Giles v Sarony seems dispositive to me.

    Has there ever been a 106(4) infringement claim involving performance of 4’33”? I’d think that a 106(1) reproduction claim would be viable, as would a Lanham Act claim, but I’d be skeptical about performance.

  • [Avatar for Anon]
    Anon
    February 26, 2023 10:46 am

    Copyright :in the prompt” is meaningless.

    “Method of provoking” IS a part of the system – prior to any choosing of inputs.

    IF you want to resolve back to outputs, then the AI is not merely “a tool,” and you are still stuck with the Naruto case.

    All of your roads lead to the same place.

  • [Avatar for Alan Harrison]
    Alan Harrison
    February 26, 2023 08:33 am

    On the other hand, the prompt to the AI could be considered a method for provoking the AI to produce art. We know that bare methods aren’t suitable for copyright. (Imagine a photographer trying to copyright their lens choice, f-stop, etc. for a particular photo.) Then it seems that the only way to recognize the AI-using artist’s creativity would be by providing them copyright in the output of their minimal effort, just as the Pollock estate may hold copyright in his splatter art.

    A very thoughtful and thought-provoking piece by Mr. Lindberg.

  • [Avatar for Alan Harrison]
    Alan Harrison
    February 26, 2023 08:29 am

    The analogies to photography, Jackson Pollock, and 4’33” are illuminating and actually shifted my perspective on this issue. While I previously felt that copyright in AI works should be prohibited as a normative policy matter (promoting the actions of human authors), reframing the question of when the creative input occurs suggests that the human artist should at least possess copyright in the “prompt” that produced the AI output.

  • [Avatar for Anon]
    Anon
    February 25, 2023 12:47 pm

    This is a setback for all the artists that would like to use artificial intelligence (AI) tools as part of their creative process.

    This begs the desired conclusion to be reached, making AI a mere “tool being used” as opposed to the ‘entity’ that derives the result of creativity.

    An error in the first instance.