New Issue: Science’s Impossible Questions. Read Now

A Compendium of AI-Composed Pop Songs

This month, in my Scientific American column, I wrote about the rise of music and art created by artificial intelligence software. What is art’s value when there’s no longer any effort involved in its creation, or scarcity to limit its ownership?

It’s time to start asking these questions, because already, people are using AI software to compose music for them. I thought I’d take a listen to some examples of AI-composed pop songs to see how far along we are.

• “Daddy’s Car.” Sony’s AI lab has created software called FlowComposer that’s capable of producing a song whose style matches whatever songs you’ve fed into it. (Here are some of its jazz snippets.)


On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.


In this case, researchers fed it Beatles songs—and indeed, the result sounds a little Beatles-ish. Note that FlowComposer produced only a lead sheet (that is, a piece of music showing the melody and chord symbols); a human then arranged it for instruments, wrote lyrics, recorded it, and mixed it. There are some fresh and intriguing chord sequences in this song, but also some noodling, aimless ones that don’t really land.

• “Mr.Shadow.” This song, too, was written by FlowComposer, but this time, it was fed songs by American songwriters Irving Berlin, Duke Ellington, George Gershwin and Cole Porter. The recording is so processed electronically that it’s a little hard to assess the composition itself, but once again, the progressions feel fairly random. Unlike an actual song by one of those composers, this one is unlikely to get stuck in your head.

• “Breaking Free.” YouTube singer Taryn Southern wrote the melody and the lyrics to this pop song, making them fit a track composed, “played,” and mixed by an AI background-music service called Amper Music. The result is much catchier and more harmonically coherent than the Sony entries, no doubt because a human did so much more of the work.

In all of these cases, even the most cutting-edge AI can’t yet compose a decent pop song without human help—the more, the better. But these are just the earliest attempts at AI composing software; the tune may be quite different in a few years.

David Pogue is the anchor columnist for Yahoo Tech and host of several NOVA miniseries on PBS.

More by David Pogue
Scientific American Magazine Vol 318 Issue 2This article was published with the title “A Compendium of AI-Composed Pop Songs” in Scientific American Magazine Vol. 318 No. 2 ()
doi:10.1038/scientificamerican022018-3BQEX9r1bDE6ojDe3Rocnc

Subscribe to Support Independent Journalism

Great science journalism requires human expertise, time, effort and creativity. And it costs money. That’s why I and the journalists here at Scientific American hope you’ll join our community.

When you subscribe, you are supporting staff and freelance journalists who are passionate about telling science stories that are true, important and compelling. Our editors and reporters are often experts in their fields, which means they understand the nuances of big discoveries and can untangle the breakthroughs from the hype. With a subscription, you are also supporting rigorous fact-checking to ensure the words we publish are precise and accurate. And you’re supporting original illustrations, graphics and photos that bring you closer to an advanced laboratory, an ice sheet in Antarctica or a space mission in orbit. You’re helping us craft other types of high-quality journalism as well: Our newsletters are carefully written, edited and curated by staffers you have or will come to know and love. Our Science Quickly podcast is based on original reporting, collaboration with editors and scientists and exacting production.

Subscriptions keep this engine running so we can continue to deliver thoughtful, rigorous and independent science journalism to you. In an era of viral misinformation, this work is crucial. If you value what we do, I hope you’ll consider joining us as a subscriber

Thank you,

Jeanna Bryner, Editor in Chief, Scientific American

Subscribe