You may be hooked on your algorithm-generated Discover Weekly playlist listening on Spotify, but what if machine learning could auto-generate music tailored to your exact taste and mood? That's what local billionaire and Khosla Ventures founder Vinod Khosla thinks will be happening within 10 years.

It sounds kind of like a scene of dialogue from HBO's Silicon Valley, but as TechCrunch reports, Khosla made comments to this effect during a recent fireside chat at Creative Destruction Lab’s second annual Super Session event. "I actually think 10 years from now, you won’t be listening to music," Khosla apparently said, though by "music" I guess he means human-created music.

It's a bold statement — and definitely one from the mind of an engineer — to suggest that computers will replace songwriting within a decade. Perhaps by "you" he just meant this audience of tech folk with whom he shares an affinity — people who may not actually care about art and meaning, per se, but are thrilled at the idea of a computer-generated soundtrack of sounds to get them through their commutes.

(Sidebar: You may recall that Khosla, from his mansion on the Peninsula, spent years battling the state of California over allowing public access to the beach beside his property near Half Moon Bay, which he was ultimately ordered to do in 2017.)

Tech writer Stuart Dredge, in a February essay on Medium, draws a parallel between the arguments being had about AI-made music today and the arguments musicians were having about synthesizers in the early 1980s, and whether they could possibly make "real" music.

Dredge writes, "While an A.I. may not be able to out-Adele Adele (or Aretha Franklin, or Joni Mitchell) with a timeless song and performance, it can compose a compelling melody for a YouTube video, mobile game, or elevator journey faster, cheaper, and almost as well as a human equivalent." And in this mode, machines are apparently getting better at making passable, listenable music.

He points to one Australian startup called PopGun, which posted this demo to YouTube about how its software can predict what may come next from a human playing a keyboard, and fit in music that follows in a similar key, pattern, and style.

Now, this is still not going to convince anyone that we won't need humans to make music anymore. But Dredge suggests that similar startups are pitching themselves to people who are already writing music, saying that AI tools can simply be aids to the creative process. One such company, Amadeus Code, says in its pitch, "Get unstuck with your songwriting with the power of artificial intelligence and say goodbye to writer’s block for good."

And here's an example of another app called ALYSIA that helps write and then sing songs. And holy shit is it bad!!

Khosla apparently has faith that the computers are just going to get better at being creative, but plenty of experts in AI have discussed the limits of it — and let's just say I fall in the camp that this is one of them. Even in 50 or 100 years I don't think we'll see machines replacing humans when it comes to writing great songs. But for sure we'll probably seeing more second-rate DJs using these tools to "make" music during their laptop sets.