Speaking of Interesting!
AI music is on a collision course
This article was originally published in the
So reads the title of one of my favorite pieces of tech commentary from last fall, published by the team at Euclid Ventures. Analyzing investment data from several sources including Carta, Pitchbook, and AngelList, the authors came to the conclusion that venture capital — an industry that, in their words, “has long celebrated itself as the business of contrarianism” — is actually becoming more consensus-driven than ever before. Investors are increasingly “outsourcing conviction” to a narrow set of proxies, like elite credentials, co-signs from incumbent accelerators, and the perception that a given market might be winner-take-all. As a result, capital is consolidating into “the same handful of themes, founders, and even funds.”
I’m afraid we’re reaching a similar point with AI music.
Since last summer, I’ve noticed that AI music activity has started converging on the same, narrow set of bets. The convergence is happening from multiple directions. AI music products are starting to look the same, in the race to own the entire music creation workflow from start to finish. The same rights holders are licensing their catalogs to AI companies, primarily for the purpose of remixing existing IP. And a crowded market of detection and attribution tools is emerging to define what counts as AI music, and who gets paid for it. (In a poetic way, the last few months alone have seen the launch of more AI music detection tools than creation tools!)
The market is incredibly competitive right now. Water & Music's free AI music database now lists over 300 apps. Against this backdrop, every stakeholder in the ecosystem faces their own distinct pressure. AI music companies need more defensible moats against the hundreds of others; labels and rights holders need to protect their catalog value; and platforms are trying to manage the integrity of their user experience amidst an influx of AI-generated content.
In nascent markets like this where the winning model is unclear, rational actors often hedge by converging on what appears to be working. The irony, of course, is that differentiation would theoretically serve these actors better.
Below, I break down three main convergences that have emerged across more than 40 AI music developments since late June 2025, and what they mean about where the market might be headed. If you care about a healthy future for the music business, you should see this AI convergence as a huge risk — of homogenization, of power concentration, and of a narrowing of what music culture could become.
(A note on scope: I’ve mostly excluded ongoing lawsuits from this analysis. My focus instead is on concrete market movements like deals being signed, partnerships being announced, and products being shipped.)
No comments:
Post a Comment