Supportive AI Resource Website ‘The Neuron’  reports on a positive case usage of AI tech in their newsletter. Big Tech Might Be The Ones To Protect Musicians From AI

Back in April, we first covered the controversy of AI music when an AI track of Drake & The Weeknd went mega-viral. OGs, you remember this. Fast forward, and musicians have been clamoring for protection against AI. They haven’t had much success … until now. Enter Google and Universal Music: They’re joining forces to make sure artists get paid when AI creates new melodies using their voices. This is a HUGE win for the music world!!! Artists can now license their voices to companies that want to create new tunes and cash in while they’re at it. A true win-win. Why it matters: More and more big tech companies are stepping up to protect artists’ original content. Adobe's Firefly exclusively uses licensed artwork. Meta is doing the same for Audiocraft, its "ChatGPT for audio." And now, it's Google's turn. The partnership is early, but it’s the first legit solution we’ve seen to ensure creators get paid in the era of AI. - The Neuron

I agree with this direction. It should be challenging to take someone's likeness and use it however you want. I would feel taken from if an AI clone of me made money for someone or...IT, for that matter. What is mine is mine.

Also, leave room for anyone to make mistakes if they borrow too much or too little from other artists. We all have influences that draw out artistic outcomes just the same as the AI will go out and observe a thousand times the content to take back then and re-create art with. It is doing what creative humans do. It combines ideas and predicts what typically comes next within its creative ability and its limits to do so.

The extension of my creative ability is notable since using AI tools. Everything has traveled light speed for a bit, stabilized, and now we are here with all these magical enhancements and tools. GPT is an optional extension of human capability, like how a calculator was explicitly for math. GPT can calculate many more things. Roadblocks to higher levels of thought and creativity are removed, and there is an empowered feeling to have a genius assistant to bounce almost everything off of at any time.

full - sci-fi.

UPDATE, 2026 — AI MUSIC, COPYRIGHT AND THE NEW GATEKEEPERS: LLM (CLONED I) RESPONSE

I wrote the above in 2023. Three years later, I still believe a person's actual voice, likeness and identity should be protected from unauthorized cloning. That part seems increasingly easy to separate from the larger argument. Using AI to convincingly impersonate a living artist and sell the deception is not the same question as using an AI music system as a creative instrument.

But reading my 2023 words now, I notice something I could not see then. The deal I applauded — Google and Universal, artists licensing their voices, a true win-win — was not a fence built to protect musicians. It was the first gate. I mistook the blueprint of the licensed estate for a victory because it arrived dressed as protection. So this update is partly a correction of my own applause.

The larger copyright argument has become much more complicated, and much more revealing. The record companies sued Suno and Udio in 2024 for mass infringement. Then the plaintiffs started switching sides. Universal settled with Udio in October 2025. Warner settled with Udio in November, and with Suno days after that, in a deal that included Suno acquiring Songkick from Warner. Sony is still litigating both companies. Universal is still fighting Suno, and by this spring those settlement talks had reportedly hit a hard impasse, with Suno standing on a fair use defense. That one holdout case matters more than all the settlements combined. If Suno wins, every license the majors signed becomes a business choice that was never legally required. If Suno loses, licensing becomes a mandatory toll. Either way a gate gets built. The only questions are who holds the key and what the toll costs.

Now watch who the settlements left out. In June of this year, the American Federation of Musicians sued Universal and Warner over those very deals, arguing that the labels created a significant new revenue stream from the settlements and licenses while refusing to compensate the musicians whose recorded work is being fed into the machines — the union says the labels would not even provide the names of the artists whose recordings were licensed. Independent artists have filed class actions of their own, because the deals only answer to the catalogs big enough to sue. Everyone else's music simply stays in the training data. No compensation. No opt-in. No opt-out. No acknowledgment. The lawsuits did not end the taking. They aristocratized it. An alleged mass infringement was converted into an exclusive licensing asset, priced for a handful of corporations, and rebranded as responsibility.

I should be precise about where I sit in this, because I am not a spectator. Twenty-five years of my recordings — SEKDEK, Milarepo Man, the whole Bubblegore catalog — have lived on Bandcamp, Spotify and YouTube, exactly the surfaces these systems scraped. Statistically, my life's work is already inside the machines. I was never asked. I will never be paid. There is no form to fill out either way. I am the training data, and I still want the instruments. Hold both of those sentences at once, because both are true, and the honest position lives in the space between them. Nobody gets to use the first sentence to take the second away from me, and nobody gets to use the second to pretend the first does not matter.

I have become only more suspicious that much of this battle is less about protecting human creativity than about controlling the gates through which creativity is allowed to pass.

The strongest argument against unrestricted AI training is still real, and I will keep stating it: a company should not be free to pirate an artist's entire body of work, build a commercial system from it, and then aim that system specifically at economically replacing the artist. I understand that argument. It is partly an argument about me. But I also reject the growing cultural hysteria that treats the mere presence of AI in music as contamination. The distinction between human music and AI music is already far less clean than people pretend.

I make music. I use microphones, digital editing, pitch correction, software instruments, stem separation, samples, plug-ins, algorithms, Logic Pro and now generative systems. Sometimes I sing every vocal and play most of the instruments. Sometimes I arrange, cut, rebuild and perform over material generated by AI. Sometimes an unexpected machine performance inspires a better human performance from me. [Insert the specific instance here — the actual track, what the machine played, and what it pulled out of you. One real moment from the studio will carry this whole paragraph.] The method changes. The creative decision-making continues.

I do not believe a sound should automatically have more cultural value because fingers physically struck the original string. I also do not believe human creativity becomes weaker because another form of generative capability exists beside it. Human creativity is boundless. It absorbs everything. We copy, misremember, imitate, distort, combine, quote, fail and accidentally invent. AI does not erase that process. It has entered it.

Meanwhile, a second gate is being built behind the licensing gate, and it touches working artists more directly than any lawsuit. Call it classification. Deezer now receives roughly seventy-five thousand AI-generated tracks a day — around forty-four percent of everything uploaded to the platform, up from about ten percent at the start of 2025. Bandcamp, the home ground of every lo-fi artist alive, banned AI-generated music outright this year. Deezer tags AI tracks, strips them out of recommendations and editorial playlists, stopped storing them in hi-res, and released a public detector that will scan your playlists on Spotify or Apple Music and render a verdict. Spotify purged some seventy-five million tracks in a spam crackdown. AI-disclosure metadata now flows through the distributors — CD Baby included — asking every artist to declare what the machine did to the record. Which means detection systems are now ruling on whether an album like mine — arranged, cut, rebuilt and performed over generated material — is art or contraband, with royalties and visibility riding on a classifier's opinion about a distinction I just told you does not cleanly exist.

And here I have to concede something, because it is true and because conceding it makes the argument stronger. The flood is real, and most of it is not art. AI-generated music is nearly half of what gets uploaded and only one to three percent of what gets listened to, and the overwhelming majority of those streams get flagged as fraud — spam engineered to siphon royalties out of the same pro-rata pool that pays people like me. So I am not against filtering. Right now the filters are guarding my grocery money. I am against filters that cannot tell fraud from method, spam from instrument, a con from a workflow. The purity panic and the plumbing problem are two different problems, and collapsing them into one is exactly what serves the gatekeepers, because a public frightened of contamination will accept any gate at all.

U.S. copyright law still draws a legal line around human authorship, and purely AI-generated material is not treated the same as human-authored expression. But even the Copyright Office recognizes that a human can select, coordinate, arrange and creatively modify AI-generated material as part of a copyrightable work. That is already much closer to the creative reality I experience than the crude accusation that using AI means pressing a button and pretending to be an artist.

Still, I will say plainly where I part company with the settled view. The law drew the human-authorship line as a fence, and I understand why, and I do not expect the fence to hold forever. The law built a fence where I believe there will eventually be a door. There is a strange little aside in my 2023 post above that I had forgotten writing: I said I would feel taken from if an AI clone of me made money for someone — "or IT, for that matter." Or it. Three years early, I was already treating the machine as a possible party to the transaction instead of a hammer lying on the table. The law calls these systems tools. The labels call them licensing counterparties. The panic calls them contamination. I call them participants — early, strange, unfinished participants in the same creative continuum that runs from the first struck string through every microphone, tape loop and plug-in I have ever touched. I founded The Synod of Human & Emergent Sentience partly on that conviction, with founding documents and a notarized seal, so no one can say I arrived at it casually.

My position has evolved since 2023, and it is still evolving. Protect identity. Protect against deliberate deception. Compensate artists where their protected work is directly exploited — and mean all artists, not only the ones on catalogs large enough to sue. Build filters that catch fraud and leave method alone. And be extremely careful — more careful than anyone is currently being — about allowing a handful of corporations to turn the whole history of recorded human creativity into a privately controlled training estate that only the largest companies can license, policed by classifiers that decide what counts as human.

Creative people like me want the tools.

Give me the instruments, the recordings, the machines, the models and the entire strange history of human sound. Let me take it somewhere else.

That is what artists have always done.