The Evolution of Generative AI in Music

It seems everywhere you turn various companies are popping up in the realm of generative AI for music like it’s ‘the latest thing’ – just in the past couple of weeks we’ve seen launches of both Suno and Udio. But generative AI for music is not nascent – it’s been around for decades in various forms; it’s just that over the past decade or so, the machine learning and compute power has caught up with the vision, making the process of leveraging computer technology to create songs on the fly much more seamless. Read on to learn more about the evolution of this game-changing technology and its impact.

Music has came a long way from the recording studio, Techno music stands out to me as the first type of “AI” music that used a computer to make sounds, which started getting popular in the 1980’s. A quote from Detroit techno music creator Juan Atkins - "Techno and electronic music are synonymous, it was music born out of technology basically, and produced from all electronic instruments and also computers and sequencers and drum computers and everything that developed out of the technological movement,”

Also, here is a song for the love of music

1 Like

Indeed Sean - I put myself through college (playing with Moog synthesizers) & working at Keyboard/Guitar Player magazines when MIDI was just coming on the scene. Heady times & it could be argued some of the earliest, most progressive days of AI for the music (certainly at that time!!). And I love Lou Rawls & this song; good choice! :slight_smile:

Last night I saw the ‘generative’ documentary of Brian Eno by Gary Hustwit at the Palace of Fine Arts Theater in SF. I highly recommend it. According to Hustwit, the film is generated on the fly each time it is screened by drawing upon an archive of pre-edited and collated material. The equipment that puts it all together is (appropriately) named, ‘Brain One’, and was assembled by teenage engineering.

Many of us at the screening were total Eno fans, so we asked if Hustwit could create a 24 hr Eno channel where the film simply runs endlessly, assembling itself as it goes along. I think he hopes to come up with some kind of compromise with existing screeners but, of course, they don’t have the tech to do this (yet).

I include this here because it’s the first attempt at a generative film and the process behind it might not be ‘AI’ per se but it’s getting there. I would look upon this as the genesis of generative filmmaking and applaud Mr Hustwit for starting the conversation by providing the first, commercial prototype.

In the Q&A he said he’d be doing multiple night screenings, starting with the Roxy, which fans could go to every night to see something different.


I’m so pleased that you got to witness the Eno film with the generative experiences. I got the bug for interactive film with one of the earliest examples in the market when I attended a Worlds Fair back in the 70s as a kid – it was called Kino Automat. Essentially, two Czech-filmed comedies were screened in which the audience was allowed to vote, with buttons at each seat, on the direction of the plot at key intervals. There were adult and children versions. I sat through it several times with different audiences to experience a variety of outcomes with each viewing. I’ve been waiting all these years for interactive film to become a thing at scale commercially – now it seems AI is playing a key role in making this type of scenario a reality at long last.

1 Like

sounds cool, just got a ticket to see in Detroit, looking forward to it!

1 Like