The U.S. Copyright Office (USCO) has issued guidance clarifying that material created solely by artificial intelligence (A.I.) cannot be copyrighted. Under the new rule, though applicants may claim a copyright for an arrangement or editing of such material, the original work is ineligible. "In these cases, copyright will only protect the human-authored aspects of the work, which are 'independent of' and do 'not affect' the copyright status of the AI-generated material itself," the USCO says.
The USCO's new rule closely follows the agency's decision to revoke a copyright granted to artist Kristina Kashtanova for A.I.-generated images in a graphic novel, reported by Reason's Joe Lancaster. Kashtanova retains copyright privileges over the work's text and the arrangement of the images, though not over the images themselves.
Generative A.I. such as the chatbot ChatGPT and image maker DALL-E can create artistic content in response to human prompts. A prompted A.I. creation does not fit traditional legal conceptions of expression and property rights.
"In the case of works containing AI-generated material, the Office will consider whether the AI contributions are the result of 'mechanical reproduction' or instead of an author's own original mental conception, to which [the author] gave visible form,'" the USCO states. "The answer will depend on the circumstances, particularly how the AI tool operates and how it was used to create the final work." The USCO recognizes the novel legal ground it treads, adding that "this is necessarily a case-by-case inquiry."
The new rule is explicit, however, that content generated by A.I. in response to prompts is ineligible for copyright. "For example, when an AI technology receives solely a prompt from a human and produces complex written, visual, or musical works in response, the 'traditional elements of authorship' are determined and executed by the technology—not the human user," it states.
"These guidelines only stress the difficulty in applying conceptions of authorship from a prior industrial era to this 21st century challenge," Mathew Dryhurst, an artist and the co-founder of spawning.ai, tells Reason, adding that a system reliant on applicants' self-reporting of A.I. authorship "seems a little frail." He questions the USCO's ability "to scrutinize which parts were created by [a machine learning] system, particularly as the generated output becomes less and less discernibly synthetic with every new model release."
Dryhurst says that "rather than avoiding the co-authorship question," we should "encourage artists and writers to experiment with offering their unique models with assurances that such a step could be economically beneficial for them and legally clear for the people who used the model." When publishing his own voice model, Holly+, Dryhurst implemented "economic incentives set to encourage those using the Holly+ model to cite Holly as a co-collaborator, and cut her into any potential profits made from her voice."
The Austrian economist Joseph Schumpeter wrote that innovation, the hallmark of capitalism, "incessantly revolutionizes the economic structure from within, incessantly destroying the old one, incessantly creating a new one." Such a force is generative A.I.; it likely requires similarly innovative policy solutions.
The post Feds Say A.I.-Generated Art Is Ineligible for Copyright appeared first on Reason.com.
