Should stars register their likenesses? – Hollywood Reporter

AI News


Metaphysic, a generative AI tech company known for popularizing a deepfake TikTok account impersonating Tom Cruise, highlighting the growing debate about owning one’s likeness in the age of AI and deepfakes. CEO Tom Graham has submitted his AI likeness for copyright registration. United States Copyright Office. He seeks to demonstrate how copyright law can be used to protect a person’s likeness from unauthorized creation and dissemination of content reproduced by his AI.

As Hollywood grapples with ethical and legal issues surrounding its use of the technology, the U.S. Copyright Office confirmed in March that most AI-generated works are not copyrightable. clarified that it is subject to protection in certain cases.

At a panel discussion during CES in January, SAG-AFTRA Duncan Crabtree-Ireland’s National Executive Director and Chief Negotiator emphasized the importance of addressing this issue. The Guild reported that he is working on policies aimed at how to create opportunities, not take them away.

In the United States, copyright law does not protect works created solely by machines. Various courts have ruled that copyright can only be granted to works created by humans. Applications that include AI-generated material can only support claims if they meet the requirement that humans contributed the majority of the work. Original work by the author. ”

In the copyright application, Graham lists himself as the author of the work. This is a crucial difference from his copyright filing by Stephen Thaler, CEO of neural network company Imagine Engines. The Copyright Office refused registration, saying that the work “lacked the necessary human copyright to support the copyright claim,” and that “the connection between the human mind and creative expression” was the subject of protection. judged to be an important factor. Thaler sued the office last year, challenging the dismissal.

Unlike Thaler’s petition and other petitions seeking protection for AI-powered works, Graham emphasizes his contribution to the creation of the video for which he seeks copyright. He says he curated the dataset used to train the model, “adding other artificial work” and using metaphysical AI tools to successfully generate realistic images of himself. increase. He points out that there was also some editing and compositing work along the way.

“I think all of these steps are very intentional, human-driven activities that are indistinguishable from someone using Photoshop or VFX tools,” says Graham. “We use AI, but it takes a fair amount of human work and human effort. It’s basically me from top to bottom.” Metaphysic develops generative AI tools and services for talent, including through partnerships with CAA.

The viability of copyright depends heavily on how the AI ​​tools work and how they were used to create the final work. For example, if an AI technician only receives prompts to create a work, the traditional elements of human copyright are determined and enforced by technology and are not subject to protection.

The standard for copyright protection is still undecided. Suzanne Hengl, an intellectual property attorney who does not represent Metaphysic, is skeptical that Graham’s application meets the agency’s current standards for human authorship. “As long as an AI tool has extensive training to work and create output, it can’t solve the problem,” she says.

A partner at law firm Baker Botts said the video could be protected if humans completed the work after it was produced. She points to the Copyright Office’s guidance that artists who use Photoshop to edit images are entitled to protection for images that have been altered.

Graham’s video may not be copyrighted, but the chief executive’s application highlights the lack of protection against illegal dissemination of videos and photographs in which the likeness of a person has been reproduced by AI. I’m here. There is no copyright law specifically designed to combat the use of deepfakes. In fact, we do allow it in some cases.

Some instances of deepfakes in Hollywood fall under the “fair use” exception to copyright infringement. This doctrine permits unauthorized use of copyrighted works in certain circumstances, such as commentary, criticism, and news reporting. One of the four factors in whether a work is eligible is the purpose and nature of the use. This extends protection for transformative works when the purpose of the copyrighted work is changed to create something with a new meaning or message.

When Kendrick Lamar used deepfakes to transform himself into Will Smith, Jussie Smollett, and Kobe Bryant in the music video for “The Heart Part 5,” the fair use exception to copyright law provided a legally sound foundation. That’s why. Although they did not agree to appear in the video, deepfakes could be seen as a transformative use to explore black representation in Hollywood, among other themes in music videos. Yes. They may also have been classified as parodies, which are also protected as transformative uses under copyright law.

One of the reasons Graham asked for AI-generated videos of himself to be registered is that unauthorized content created using AI tools, such as deepfake porn, can help people identify, perform, and brand. in order to protect Under the Digital Millennium Copyright Act, copyright owners can promptly remove material that infringes their copyright.

Metaphysic Attorney Jordan Manekin said: “This is a much easier remedy for most people than going to court and spending years trying to get it to work.”

But even if protection were granted, it’s unlikely Graham would be able to use the copyright to remove the AI-generated photos and videos himself, Hengl said. “In that copyright can be used as a way of enforcing likeness rights or enforcing privacy rights, it’s trying to fit a square peg into a round hole,” she suggests. “Protection can only be enforced against works that are substantially similar. There must be more than an image of the person — clothing, facial expressions, body pose, background — that is similar.”

Hengl said that traditional avenues of action against deepfakes and similar technologies have pursued allegations against violations of publicity and privacy rights.

As SAG-AFTRA’s Crabetree-Ireland summarized at CES, “As this technology democratizes and becomes more accessible, this will become a broader public policy issue.”





Source link

Leave a Reply

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