Fearing copyright issues, Getty Images bans AI-generated artwork

A selection of stable diffusion images with a strikethrough across them.
Zoom in / A selection of stable diffusion images with a strikethrough across them.

Ars Technique

Getty Images has banned the sale of AI generative artwork created using image synthesis models such as Stable Diffusion, DALL-E 2, and Midjourney through its service, The Verge reports.

To clarify the new policy, The Verge spoke with Getty Images CEO Craig Peters. “There are real concerns about the copyright of the outputs of these models and unaddressed rights issues with respect to the images, image metadata and individuals contained in the images,” Peters told the publication.

Getty Images is a large archive of archival and archival photographs and illustrations, often used by publications (such as Ars Technica) to illustrate articles after paying a fee.

Getty’s move follows image synthesis bans by smaller art community sites earlier this month, which found their sites flooded with AI-generated work that threatened to overwhelm the artwork. created without the use of those tools. Shutterstock, a competitor of Getty Images, allows the creation of AI-generated artwork on its site (and although Vice recently reported that the site was removing AI artwork, we still see the same amount. and Shutterstock’s submission terms have not changed).

A notice from Getty Images and iStock about a ban
Zoom in / A notice from Getty Images and iStock about a ban on “AI-generated content”.

Getty Images

The ability to protect AI-generated artwork has not been tested in court, and the ethics of using artists’ work without consent (including artwork found on Getty Images) to train neural networks in able to create works of art on a near-human level is still an open question under discussion online. To protect the company’s brand and its customers, Getty has decided to avoid the problem with its ban altogether. That said, Ars Technica searched the Getty Images library and found AI-generated artwork.

Can AI artworks be copyrighted?

While the creators of the popular AI image synthesis models insist that their products create copyrighted works, the copyright issue on AI-generated images has not yet been fully resolved. It is worth pointing out that an often cited article in the Smithsonian titled “US Copyright Office Rules AI Art Can’t Be Copyrighted” is misspelled and is often misunderstood. In that case, a researcher attempted to register an artificial intelligence algorithm as a non-human owner of a copyright, which the Copyright Office denied. The copyright owner must be human (or a group of human beings, in the case of a company).

Currently, AI image synthesis companies operate on the assumption that copyright for AI artwork can be registered to a human or company, just as it is with the output of any other art tool. There is strong precedent in this, and in the Copyright Office’s 2022 decision rejecting copyright registry to an AI (as mentioned above), it referred to a historic lawsuit from 1884 claiming the state copyright of the photographs.

At the beginning of the camera story, the defendant in the case (Burrow-Giles Lithographic Co. v. Saronia) He claimed that the photographs could not be protected by copyright because a photo is “a reproduction on paper of the exact characteristics of a natural object or a person”. In fact, they argued that a photo is the work of a machine and not a creative expression. Instead, the court ruled that the photos may be copyrighted because they are “representatives of original intellectual conceptions of [an] author.”

People familiar with the AI ​​generative art process as it is now, at least as far as text-to-image generators are concerned, will recognize that their image synthesis results are “representatives of original intellectual conceptions of [an] author. “Despite misconceptions to the contrary, the creative input and guidance of a human being is still required to create an image synthesis work, no matter how small the contribution is. Even the selection of the tool and the decision to execute it. it is a creative act.

Under U.S. copyright law, pressing the shutter button of a randomly pointed camera at a wall still assigns the copyright to the human who took the photo, yet human creative input into a work of art image synthesis can be much broader. So it would make sense if the person who initiated the AI-generated work owns the copyright of the image unless otherwise limited by the license or terms of use.

That said, the copyright issue on AI artwork has yet to be resolved legally one way or another in the United States. Stay tuned for further developments.