NYT critic Jon Caramanica: AI in music is the new Auto Tune — already in every session, widely accepted within 24 months
Jan 22, 2026 with Jon Caramanica
Key Points
- Jon Caramanica argues AI is already embedded in professional music production and will reach mainstream cultural acceptance within 12 to 24 months, following the historical arc of Auto Tune's normalization.
- Junior studio staff in Los Angeles songwriting sessions are deploying AI tools to reshape workflow in ways inaudible to listeners, with artist Teddy Swims confirming public use of AI for vocal multitracking and key-change previews.
- Caramanica's deeper concern is passive consumption accelerated by AI-generated background content, which marginalizes artists whose work rewards close attention rather than ambient listening on playlists.
Summary
Jon Caramanica, pop music critic at the New York Times and co-host of the Popcast, argues that AI is already embedded in professional music production and that widespread cultural acceptance is roughly 12 to 24 months away.
The mechanics are already in motion. In any major Los Angeles songwriting session, the room typically layers a featured artist, their personal collaborator, a name producer, and one or two lower-level studio hands. Caramanica's position is that those junior participants are already deploying AI tools in ways that are not audible to the listener but are reshaping workflow. Teddy Swims confirmed as much publicly, describing using AI to multitrack vocals and preview key changes before committing to a take.
The historical parallel Caramanica reaches for is Auto Tune. T-Pain normalized pitch-correction software at commercial scale, Jay-Z declared its death in 2009 with "D.O.A.", and today Auto Tune appears across country, hip-hop, pop, and reggaeton without serious critical resistance. The Dylan-goes-electric moment at Newport carries the same structural logic: technology that triggers anxious discourse at emergence becomes invisible infrastructure within a generation. AI in music is on the same arc, just compressed.
The more nuanced concern Caramanica raises is not the obviously synthetic artist, citing Solomon Ray as an example of a purported solo act likely constructed by running Anthony Hamilton recordings through generative software. He expects platforms like Spotify and Apple to resolve that through labeling or revenue reallocation, in the same way sampling litigation eventually created formal licensing frameworks. What concerns him more is a behavioral shift already underway before AI arrived: the majority of listeners treat music as ambient, pressing play on a playlist and disengaging. AI accelerates the production of that background-ready content but did not originate the demand for it. The downstream risk is that passive consumption becomes the dominant mode, further marginalizing artists whose work rewards close attention.