Last month, a British indie band called Breezer made headlines worldwide for releasing a 'lost Oasis album', or more accurately, what they called an “alternate reality concept album”, created by AI in the style of the [still not reunited, at the time of writing] Manchester Britpop superstars. The 'Aisis' album even attracted praise from the famously acerbic Liam Gallagher, who declared it "mad as fuck" on the basis of the sole track he listened to, adding “It’s better than all the other snizzle out there".
Not everyone is so enthused about the increasingly popular trend of creating AI music in the style of hallowed artists.
According to an April 11 story in the Financial Times [paywalled] Universal Music Group, which owns the rights to approximately one third of the global music market, has requested that leading streaming platforms, including Spotify, block AI services from accessing and 'scraping' melodies and lyrics from copyrighted material.
And it's not just the music industry 'suits' who're alarmed. Earlier this year Nick Cave lamented the use of ChatGPT to create songs ‘in the style of Nick Cave’ calling the results "bullshit" and "a grotesque mockery of what it is to be human."
And yet, it's hard to deny that, on occasion, the results can be impressive. A recently uploaded AI creation seeking to imagine The Beatles covering The Beach Boys classic God Only Knows, one of Paul McCartney's favourite songs, is rather cute, while the work of YouTuber FaustoX - most of which focus on an AI-generated Nirvana frontman Kurt Cobain - is striking, not least the video which sees an AI Lana Del Rey covering Nine Inch Nails' Hurt in a similar style to Johnny Cash's iconic take on Trent Reznor's bleak masterpiece is both exquisitely beautiful and slightly creepy.
Beneath FaustoX's videos, the comment sections are full of impressed music fans requesting more such creations. In the closing years of the last century, the music industry was terrified of the threat posed to its bottom line by file-sharing service Napster: it may have seen nothing yet.
