When, if at all, can artificial intelligence be considered “human?” Who is responsible for the art created by technology? Who owns art primarily created by computers? The US Copyright Office addressed these issues in its latest decision regarding artificial intelligence which will have implications for art and NFTs in the future.
In short, the Copyright Office has ruled that it will not offer protection if it determines that a human being did not create a work of art. However, a closer look at the justification for the copyright claim itself and its subsequent denials reveals a deeper and more complex web of issues that the Copyright Office will have to grapple with in the years to come. to come.
The U.S. Copyright Office’s AI Decision
In 2019, Dr. Steven Thaler, founder and board member of Imagination Engines, Inc., attempted to copyright a two-dimensional work of art titled “A Recent Entry into Paradise” . According to Thalier, this piece is a “simulated near-death experience” in which an algorithm reprocesses images to create hallucinatory images and a fictional afterlife narrative. Critically, the computer is meant to complete this work of art with minimal human intervention.
In Thaler’s original copyright application, the author of the artwork was identified as the “Creativity Machine”, with Thaler listed as the claimant alongside a statement of transfer: “ownership of the machine”. In his application to the Copyright Office, Thaler left a note stating that the artwork “was autonomously created by a computer algorithm running on a machine” and that he was “seeking to register this work computer-generated as work for hire for the owner of the Creativity Machine. In a letter dated August 12, 2019, a Copyright Office registration specialist declined to register the claim, finding that it “lacked the human authorship necessary to support a copyright claim”.
Thaler then asked the Copyright Office to reconsider its initial refusal to register the artwork, arguing that “the requirement of human authorship is unconstitutional and not supported by statute or case law. “. The Copyright Office reassessed the claims and again concluded that the work “did not have the human authorship required to support a copyright claim” because Thaler had “provided no evidence of ‘a creative contribution or sufficient intervention of a human author in the work’. The Copyright Office went even further and said that it “would not abandon its longstanding interpretation of the Copyright Act, the Supreme Court, and the judicial precedent of lower courts according to which a work meets the legal and formal requirements of copyright protection only if it is created by a human author.
Thaler then submitted a second request for reconsideration, again arguing that the Copyright Office’s “human authorship” requirement is unconstitutional and unsupported by case law. Specifically, in this second request, Thaler argued that the Copyright Office “should” register copyrights in machine-generated works, as doing so would “pursue the underlying purposes of copyright law.” copyright, including the constitutional justification for copyright protection”. In response to the Copyright Office’s citation of case law dealing with human authorship, Thaler asserted that “no binding authority prohibits copyright for [computer-generated works]”, that copyright law already permits non-human entities to be authors under the work-for-hire doctrine, and ultimately that the Copyright Office ” is currently relying on non-binding court opinions from the Golden Age to answer the question of whether [computer-generated works] can be protected.
In response to Thaler’s second request, the Copyright Office once again ruled against Thaler. A three-person review board determined, as before, that the image created by Thaler’s AI did not include the element of “human authorship” necessary for copyright protection.
So what does this mean for computer-generated works in the future? To illustrate this forward-looking question, let’s dive into the past and take a look at a piece of cinematic art, 2014’s Ex Machina.
In the 2014 film, Ex Machina, a computer coder named Caleb (Domhnall Gleeson) is asked by Nathan (Oscar Isaac), the inventor CEO of a massive tech company, to be the “human component” in a Turing test. which will determine the abilities of a robot named Ava (Alicia Vikander). In this film, Nathan creates the code to build Ava, and Caleb helps test Ava’s abilities. In one scene, however, Ava creates a design for Caleb. At this point in the movie, it’s clear that Nathan created Ava, and it’s also clear that Caleb is a human element in bringing Ava to her final form. Amidst all this clarity, however, is the unanswered and unanswered question: Who owns the artwork that Ava created?
Is it Nathan, since he created the software, the code and physically built Ava? Is it Caleb, since the work was created under his direction and for his benefit? Or is it Ava, since she literally drew this piece of art? If we approach this issue as the Copyright Office has, Ava would not be eligible for ownership of this art solely on the basis that, despite her “human” features, she is, in essence, a computer. As the film suggests, we need to assess how we understand our future relationship with artificial intelligence, and part of that consideration is the impact of this Copyright Office decision on the future of NFTs.
Like the artwork created by the “Creativity Machine”, many NFT art projects are generative. Generative Art, often referred to as “Coding Art”, is a process in which a person provides base layers of art and/or various codes and algorithms which are then processed by a computer which produces artwork. The artwork used will generally be the property of the artist. The code used will generally be the property of the creator of the software. But the question remains: who owns the output of the computer?
Although there has been recent case law on the matter (see Rearden LLC v. Walt Disney Co., 293 F. Supp. 3d 963 (ND Cal. 2018) (acknowledging that some authorities “suggest that the copyright protection given to a computer program may extend to the output of the program if the program ‘does the lion’s share of the work'” )), it’s safe to say there are still complex and unresolved issues in copyright law. Perhaps the Copyright Office and other government agencies will provide additional guidance on these evolving issues.
For the full decision of the US Copyright Office Review Board, see here.
For an overview of copyrighted works, see here.