Copyright issues have dogged AI since chatbot tech gained mass appeal, whether it’s accusations of entire novels being scraped to train ChatGPT or allegations that Microsoft and GitHub’s Copilot is pilfering code.
But one thing is for sure after a ruling [PDF] by the United States District Court for the District of Columbia – AI-created works cannot be copyrighted.
You’d think this was a simple case, but it has been rumbling on for years at the hands of one Stephen Thaler, founder of Missouri neural network biz Imagination Engines, who tried to copyright artwork generated by what he calls the Creativity Machine, a computer system he owns. The piece, A Recent Entrance to Paradise, pictured below, was reproduced on page 4 of the complaint [PDF]:
The US Copyright Office refused the application because copyright laws are designed to protect human works. “The office will not register works ‘produced by a machine or mere mechanical process’ that operates ‘without any creative input or intervention from a human author’ because, under the statute, ‘a work must be created by a human being’,” the review board told Thaler’s lawyer after his second attempt was rejected last year.
This was not a satisfactory response for Thaler, who then sued the US Copyright Office and its director, Shira Perlmutter. “The agency actions here were arbitrary, capricious, an abuse of discretion and not in accordance with the law, unsupported by substantial evidence, and in excess of Defendants’ statutory authority,” the lawsuit claimed.
But handing down her ruling on Friday, Judge Beryl Howell wouldn’t budge, pointing out that “human authorship is a bedrock requirement of copyright” and “United States copyright law protects only works of human creation.”
“Non-human actors need no incentivization with the promise of exclusive rights under United States law, and copyright was therefore not designed to reach them,” she wrote.
Though she acknowledged the need for copyright to “adapt with the times,” she shut down Thaler’s pleas by arguing that copyright protection can only be sought for something that has “an originator with the capacity for intellectual, creative, or artistic labor. Must that originator be a human being to claim copyright protection? The answer is yes.”
Unsurprisingly Thaler’s legal people took an opposing view. “We strongly disagree with the district court’s decision,” University of Surrey Professor Ryan Abbott told The Register.
“In our view, the law is clear that the American public is the primary beneficiary of copyright law, and the public benefits when the generation and dissemination of new works are promoted, regardless of how those works are made. We do plan to appeal.”
This is just one legal case Thaler is involved in. Earlier this year, the US Supreme Court also refused to hear arguments that AI algorithms should be recognized by law as inventors on patent filings, once again brought by Thaler.
He sued the US Patent and Trademark Office (USPTO) in 2020 because patent applications he had filed on behalf of another of his AI systems, DABUS, were rejected. The USPTO refused to accept them as it could only consider inventions from “natural persons.”
That lawsuit was quashed then was taken to the US Court of Appeals, where it lost again. Thaler’s team finally turned to the Supreme Court, which wouldn’t give it the time of day.
When The Register asked Thaler to comment on the US Copyright Office defeat, he told us: “What can I say? There’s a storm coming.” ®
Bootnote
Thaler has been making bold claims about the promise of AI for decades. An article from a dusty 1996 copy of New Scientist headlined “The creativity machine” opens:
