At the start of this year, Saving Country Music — one of the most-read independent country outlets around — announced a new editorial policy for 2026: every artist, label, or publicist submitting music for review now has to disclose whether AI was involved in writing or producing it. Music that's majority AI-written, or that contains any audible AI-generated sound at all, gets excluded from reviews and year-end lists entirely.
I read the whole announcement, and the comment section underneath it, twice. Partly because I run a station that's built entirely around AI-assisted music, and partly because — and I mean this honestly — I think the policy is a genuinely good idea.
What the policy actually says
The rules are specific. If 50% or more of a song's lyrics were generated by AI, or if even a single audible sound in the track was AI-produced, it has to be disclosed — and the outlet won't consider it for review coverage or best-of lists. Using AI the way you'd use a thesaurus or a rhyming dictionary, to help find one word or check a fact, doesn't count. Having AI write or perform the song does.
There's a 60-day grace period before it's enforced in full, and the outlet has said clearly it isn't going to pretend AI music doesn't exist — it'll still write about the phenomenon, including huge AI hits, just not review them alongside human-made records.
Why I actually like this
This whole station exists because I believe listeners deserve to know how something was made. Every song page, every FAQ answer, the About page — it all says the same thing: I write the lyrics and, often, the melody and chords myself, and AI helps arrange, perform, and produce the finished recording. That's not a marketing angle. It's just what's true. A policy that asks every artist to say the same kind of thing, plainly, before their music gets covered, is a policy I'd sign onto without hesitation.
What stood out to me most, reading the comments underneath the announcement, wasn't the disagreement about AI itself — it was a fair question about consistency. A few people pointed out that heavily corrected vocals or backing tracks in live performances have quietly gone unlabeled for years, while a single AI-generated thunderclap in a song about a storm would now need a disclosure. That's a reasonable thing to sit with. If transparency matters, it probably shouldn't stop exactly at the point where "AI" is the word being used.
Where a station like mine actually sits
It's worth being precise here, because "AI music" gets used as one big label for very different things. Fully AI-generated acts — the kind that write nothing themselves and let a prompt do all of it — sit at one end. What happens on this station sits somewhere else: the writing is entirely mine, start to finish, and AI is doing the part I currently don't have the stamina for. Under a policy like this one, that distinction is exactly the kind of thing disclosure is supposed to surface, instead of leaving everyone to guess.
I don't think a single policy from one outlet is going to settle what AI in music should look like — that's going to take years, and probably a lot more disagreement than agreement along the way. But I'd rather have more publications, platforms, and stations drawing an honest line in public than none at all. If you're curious what that line looks like for this particular station, it hasn't really changed: read the full story here, or just go straight to the songs.