AI singer-songwriter ‘Anna Indiana’ debuted her first single ‘Betrayed by this Town’ on X, formerly Twitter—and listeners were not too impressed.
AI singer-songwriter ‘Anna Indiana’ debuted her first single ‘Betrayed by this Town’ on X, formerly Twitter—and listeners were not too impressed.
The threshold for how much human input counts as “authorship” is extremely low. Photographers get copyright over pictures they take when their sole contribution to the image is aiming the camera and pushing a button. Most AI-generated art involves a lot of human input in the form of prompting, selecting outputs, and then often tweaking or splicing them together in various ways.
But even if by some weird twist US courts do rule this sort of thing to be public domain, why wouldn’t this “threaten human artists?” Having awesome AI-generated art being public domain seems like the best of both worlds to me - you get awesome art and it’s legally unencumbered. How would a human artist compete with that? Their art would be more expensive and you’d have all kinds of limitations on what you can do with it.
If the autogenerated art becomes too close to copyrighted art, then you’ll have humans suing AI generators.
George Harrison’s My Sweet Lord is very similar to He’s So Fine by the Chiffons. And that was an easy case. But some cases in requires deeper analysis, such as Lana Del Ray’s Get Free.
https://www.bbc.com/culture/article/20190605-nine-most-notorious-copyright-cases-in-music-history
If AI is sampling, then how do you defend it being unintentional? While all Radiohead sought was credit on the writing (in this case), would humans (whose livelihood is being threatened) be so generous with an AI composition? And if the music industry is threatened by AI, they will lawyer up.