Arts·Commotion

Timbaland used an independent producer's work to train AI — but without the artist's consent

Music insiders Dylan Green and Michelle Santosuosso discuss musicians' rights in the age of AI.

Music insiders Dylan Green and Michelle Santosuosso discuss musicians' rights in the age of AI

A Black man in all white dances on a stage with headphones on and a laptop nearby and a bunch of people surrounding him
Timbaland is under fire for how he uses and trains AI. (Photo by Steve Jennings/Getty Images for Revd)

Timbaland recently used hip-hop artist K Fresh's beat to train his artificial intelligence music platform, Suno — but without K Fresh's permission. 

This isn't the first time that the music producer has gotten into hot water around using AI in his music. But he's not the only big name using the technology to enhance his music. Fans also decried Erykah Badu and the Alchemist for using AI in the cover art of their new single together. 

Today on Commotion, guest host Rad Simonpillai speaks with music journalist Dylan Green and veteran music industry insider Michelle Santosuosso to discuss the Timbaland backlash and what protections artists need in the age of AI. 

We've included some highlights below, edited for length and clarity. For the full discussion, listen and follow Commotion with Elamin Abdelmahmoud on your favourite podcast player.

WATCH | Today's episode on YouTube:

Rad: Timbaland was claiming that he's not trying to jack this original beat from this producer, he was just trying to demonstrate the ways that AI could be used to remix an already existing composition. Is that right?

Dylan: That's what he claims. But a lot of people, myself included, think it really just borders on straight-up stealing. Generative AI is a whole bag of worms that I'm sure we're going to get into here, just in the sense that this is a computer spitting back out stuff that it's already learned, as opposed to creating anything — which is very, very close to theft, at least in my opinion.

Rad: Dylan, you've written some criticisms about not just Timbaland, but other high profile artists like Erykah Badu and the Alchemist, who've all been receiving their fair share for using AI, with criticisms coming, in large part, from the hip-hop world. What are people saying?  

Dylan: People are upset that real artists are being cut out of actual financial opportunities. The fact that these are all people that can afford to put people on, essentially, and they're just not, in a way to save as much money as they can. Being a legacy artist, and a Gen Xer in particular, just trying to get in on the ground floor of what you think is going to be the most popping thing in five years, it just reeks of laziness in so many ways, in my and a lot of people in my circles' opinion. 

It starts to feel predatory after a while. Because it's a difference between being on TikTok and listening to people's music and giving criticisms, giving praise and then being like, "Hmm, I like this thing, let me see what happens if I feed it into this thing without telling them." It's an abuse of trust in a lot of ways.

Rad: Michelle, speaking of that abuse of trust, because I think it's interesting that Timbaland is getting all this heat for how he's using AI and how troubling it is from an ethical perspective. But when we talk about the legal perspective, his actions actually seem okay. And that's baffling to me. But tell me, why is that?

Michelle: It's because there's no real legislation around it. Now, using K Fresh's music without consent, that violated Suno's terms of service. But under current law, there's nothing that you need to delete if AI learned from that actual beat, from the ingestion of it. And this is the rub: music and lyrics that are 100 per cent made by AI are ineligible for copyright protection. But the legal complications, I call it the three C's: there are no laws around consent, credit or compensation for artists who generative AI is training on. So while these systems are freely training on both copyrighted material and independent artists — which is the case with this producer — until we get legislation around credit, compensation and consent to use it in the first place, this is, I agree with Dylan, it's mostly going to disenfranchise smaller artists that are trying to come up. 

Rad: Ultimately, how are you seeing these tensions — between AI and then human artists and then their fan bases and just the music industry at large — playing out? 

Michelle: It's clearly shaping up to be a battle. And unlike streaming's pro-rata royalty problem, which was very confusing for consumers, this subject has successfully hit the radar of the music consumers. 

But I want to point something out to people about artists' paycheques. I want to put it out that copyright ownership has a bundle of rights with it, that each of those rights has their own income stream attached to it, and there's five of them: to reproduce the work, to distribute the work, perform the work, make a derivative work, or display it. And each one of those has an income stream attached to it. 

So the tech oligarchy is going to continue to try to normalize copyright theft. If you look at how music has been devalued in the digital age already — and I'm saying from filesharing Napster, company that I actually worked at — it's upwards of 90 per cent from the sales area, in terms of the devaluation of music. So we have to compensate people for the artistic endeavors that they make. And that piece is not part of law. 

You can listen to the full discussion from today's show on CBC Listen or on our podcast, Commotion with Elamin Abdelmahmoud, available wherever you get your podcasts.


Panel produced by Ty Callender.

ABOUT THE AUTHOR

Sabina Wex is a writer and producer from Toronto.