Artificial Intelligence is poised to take over the world, and it's already capable of creating art. AI systems have created Beethoven-inspired compositions, written the next book for Game of Thrones, and it can even paint. However, AIs lack the one key component all other artists have -- a human body. Thus, a growing question in the realm of artificial intelligence is who owns what an AI produces?
Here's the current policy in the United States: AIs cannot be awarded copyright for their creations. The US Copyright Office does not accept claims for works not made by humans. This understanding is also how a monkey who snapped a selfie lost the rights to that selfie in a lengthy court battle. In the recent court ruling, the photographer David Slater agreed to donate 25 percent of his profits from that photo to protecting the macaques.
"PETA and David Slater agree that this case raises important, cutting-edge issues about expanding legal rights for nonhuman animals, a goal that they both support, and they will continue their respective work to achieve this goal," read a joint statement on the group's website.
As silly as that sounds, the transition from monkeys snapping pictures (or even elephants painting on canvases) to AI systems generating art isn't that far.
Annemarie Bridy works as a professor of law at the University of Idaho. She said the U.S. law doesn't technically require human authorship (although the contingencies do now specifically address both elephants and monkeys).
"Courts have always assumed that authorship is a human phenomenon," she said.
AIs make this distinction even hazier as they're created by humans. So does the one person responsible for building the software earn the copyright? What about the team responsible for hardware? Can a machine itself earn copyright protection? (That last question might also tap into ongoing debates about whether scientists can and should program emotions and expressions into a robot.)
According to the Art Law Journal, "copyright protection is available for 1) an original work of authorship, 2) fixed in a tangible medium 3) that has a minimal amount of creativity. If a work doesn’t have all three of these components, then it is not copyrightable subject matter."
And thanks to the monkey lawsuit, most of the discussion has settled. Computers (and monkeys) can't go to court. However, shouldn't someone still own the rights to what's created by an AI?
Courts in the US attempted to settle the issue in 1974 with the National Commission on New Technological Uses of Copyrighted Works (CONTU). However, it was 1974 and (shocker) there wasn't a ton of discussion surrounding authorship of computer-generated creations. The current assumption is that whoever plays an extensive role in developing the technology should earn the copyright to the creations produced by the AI. However, "extensive" has yet to be defined, and the sooner that definition comes, the better prepared the world will be.
Researcher and writer Robert Hart made an excellent point in a recent analysis posted to Quartz:
"[The products of AI] promises great benefits in the fields of science, technology, and medicine—but it does in the creative realms, too. If we don’t resolve these thorny issues of ownership now, we risk delaying the delivery of these benefits across all industries. Our laws need to adapt to the reality of the modern world, and they need to do so quickly."