US Copyright Office withdraws copyright for AI-generated comic artwork

the cover of
Expanding / The cover of the manga “Dawn of Zarya” produced using Midjourney’s AI image synthesis in 2022.

Chris Kashtanova

On Tuesday, the U.S. Copyright Office declared images created using the AI-powered Midjourney image generator to be for comic books. Dawn of Zarya No copyright protection should have been granted and the image copyright protection will be revoked.

In a letter to author Kris Kashtanova’s attorney obtained by Ars Technica, the Secretariat clarified “incomplete information” in the original copyright registration, revoking the original registration and seeking protection for AI-generated material. citing it as the reason why they plan to issue new registrations. image. Instead, the new registration covers only the text and image and text placement of the work. Initially, Kashtanova did not disclose that the images were created by his AI model.

“We conclude that Ms. Kashtanova is the author of the text of the work and the selection, coordination, and arrangement of the written and visual elements of the work,” reads the copyright letter. “Its copyright is protected by copyright. However, as explained below, the images in the work produced by Midjourney technology are not human works.”

Last September, in an article first published on Ars Technica, Kashtanova publicly announced: Dawn of Zaryacontains comic book style illustrations generated from prompts using latent diffusion AI processing was copyrighted. At that time, I thought it was a precedent for registering works by latent diffusion.

9 月に最初に米国で著作権登録を受けた AI 支援コミック <em>Zarya of the Dawn</em> Excerpts from are currently being revised.  ” src=”https://cdn.arstechnica.net/wp-content/uploads/2022/09/zarya_hero-640×358.jpg” width=”640″ height=”358″ srcset=”https://cdn.arstechnica .net/wp-content/uploads/2022/09/zarya_hero.jpg 2x”/><figcaption class=
Expanding / Excerpt from an AI-assisted comic Dawn of Zaryafirst received US copyright registration in September and is now being revised.

However, as explained in the letter, the Copyright Office gave Kashtanova a reason in October after it learned through her social media posts that the work contained AI-generated imagery. We have issued a notice that we intend to cancel your registration unless you provide additional information. Registration should not be canceled. Kashtanova’s lawyers responded to the letter in November, claiming that Kashtanova wrote all aspects of her work, with Midjourney serving merely as a supporting tool.

This argument is not sufficient for the Copyright Office, which details why it believes that artwork generated by AI should not be granted copyright protection. In a key excerpt shown below, Office highlights the machine-generated origins of the image.

Based on previous records, the Office determined that the images produced by Mid-journey included in the work is not an original work protected by copyright. Copyright. look HaSpending (Thard ) § 313.2 (explaining “the Office does not register works”) generated by machines or mere mechanical processes acting randomly or automatically without creative input or intervention from a human creator”).Although she claims she has “Guiding” the structure and content of each image, a process described in Kashtanova’s letter It reveals that it was Midjourney, not Kashtanova, that created the “traditional element”. Image author’s.

The letter provides an additional analogy for understanding why the Copyright Office considers Kashtanova not the creator of the image. This includes the idea of ​​hiring a human to create images with descriptions and performing text-based image searches on the Internet. The letter’s overall allegations could set an important legal precedent for future attempts to copyright AI-generated images.

In an Instagram post, Kashtanova responded to the letter by framing it as an overall triumph for AI augmented artists. She said the ruling is “great news” in the sense that it protects the comic’s story and image placement, and that it “covers many uses of AI art for people in her community.”

But when it comes to losing copyright protection for individual images, Kashtanova says she hasn’t given up the fight.

I was disappointed with one aspect of the decision. The Copyright Office did not agree to acknowledge my copyright on individual images. It is fundamental to understand that the output of a generative AI model is directly dependent on the artist’s creative input and is not random. My attorney is considering options to further explain to the Copyright Office how individual images created by Midjourney will be treated. [a] It is copyrighted because it is a direct expression of my creativity.

apart from Artwork precedents generated by previous algorithms Because they are copyrighted, this ruling means that AI-generated images that do not contain human-made elements cannot currently be copyrighted in the United States. Copyright Office decisions on this matter are likely to be retained unless challenged in court, amended by law, or reconsidered in the future.

This ruling may eventually be reconsidered as a result of cultural shifts in how society perceives AI-generated art. This could allow new interpretations by various members of the US Copyright Office over the next decade. For now, AI-powered artwork is still a novel and poorly understood technology, but it could eventually become the standard way visual art emerges. Denying copyright protection could preclude future use by large and powerful media conglomerates. So the AI ​​and copyright story isn’t over yet.



Source link

Leave a Reply

Your email address will not be published. Required fields are marked *