The Enshittification of Music Tools
I just read a Reddit post that left me feeling both deeply sad and, in a strange way, cautiously hopeful about the future of AI in audio and music production.
The sadness comes from the fact that we are here at all. Allegedly, a company trusted by working musicians, engineers, and producers appears to be circling the same generative-AI drain that has already distorted so much of the technology world. Another professional toolmaker, another legacy brand, another ecosystem built on trust and craft, now seemingly tempted by the promise of automation, investor excitement, and whatever flavor of “innovation” currently sounds good in a boardroom. We don’t know if any of this is true, since it was written by an employee upset about possible Gen AI integration with software, but if experience has anything to say about it, its at the very least believable.
But the hopeful part is just as important: people are not confused about this.
The response I saw was overwhelmingly hostile toward generative-AI integration in music software, but it was not anti-technology in some reactionary or simplistic way. People seem to understand the distinction between machine-learning utilities and generative-AI systems. That distinction matters, because it is often flattened by executives, marketers, and AI evangelists who would like every objection to be dismissed as fear of progress.
Most musicians and engineers are not rejecting useful tools. Stem separation, drum alignment, restoration, repair, assistive analysis, accessibility improvements, and certain forms of intelligent workflow support can all have legitimate places in modern audio work. We already use tools that would have seemed impossible a generation ago. Audio production has always involved technology, abstraction, modeling, automation, and systems that extend human ability.
But there is a line.
There is a difference between a tool that helps a person do the work and a system designed to replace, imitate, or extract from the person doing the work. There is a difference between restoration and substitution. There is a difference between analysis and authorship. There is a difference between using machine learning to solve a technical problem and using generative AI to turn human creative behavior into training material for platforms that may eventually compete with the very people who made the data valuable in the first place.
I can say this plainly: any integration with Suno or another generative-AI music company would be the end of me buying those products. I do not think I am alone in that.
I personally do not want anything to do with generative AI in music production. I do not want to promote it. I do not want to normalize it. I do not want it inside the tools I use to make records with human beings. I recently left a volunteer position because I did not even want to have to think about a generative-AI company sponsoring events I was connected to.
I understand that some people, including very visible artists and producers, want to frame this as inevitable. They want to say the writing is on the wall.
But fuck that.
I do not accept inevitability as an argument. Many terrible ideas have been sold as inevitable by people who stood to benefit from everyone else surrendering early. The fact that executives, investors, and celebrity opportunists are excited about something does not make it culturally good, ethically defensible, or creatively necessary.
The privacy issue may be even more serious. If a DAW, plugin manager, or audio software ecosystem is collecting workflow data that could be used for AI training, that is not a minor technical concern. That is a labor issue. That is a creative exploitation issue. That is a trust issue at the foundation of the relationship between toolmaker and user.
No one wants to buy tools from a company that may be studying their creative behavior in order to train systems that could eventually replace them.
This is especially galling because what working audio people actually want is not complicated. We want stability. We want fewer glitches. We want better offline usability. We want comping that works. We want accessibility. We want reliable integration. We want native plugin availability. We want professional tools that behave professionally in front of clients. We want the ecosystem to be robust enough that it does not interrupt the work it is supposed to support.
That is not a nostalgic demand. That is not anti-progress. That is the baseline expectation for tools used in serious creative labor.
What saddens me is that companies in the professional audio world have earned real loyalty through great hardware, great plugins, and years of genuinely useful engineering. That kind of trust is difficult to build. It comes from solid reliability, great sound, workflow, support, and the sense that the company understands the people who use its products. But trust can disappear very quickly when users feel that they are no longer being treated as musicians, engineers, producers, or collaborators, but as data sources, subscription targets, and raw material for the next speculative product cycle.
People keep describing this AI push as executive-class hype, investor appeasement, and enshittification. That may sound blunt, but it is difficult to argue with the pattern. Across the technology sector, companies keep forcing AI into products not because users asked for it, but because leadership wants growth optics. The result is rarely better tools. More often, it is friction, surveillance, instability, worse support, and a creeping sense that the product is being redesigned around the company’s ambitions rather than the user’s needs.
The professional audio community seems to be aligning around a very clear demand:
Keep building high-quality, human-centered tools.
Do not partner with generative-AI music companies.
Do not train AI on user data.
Do not turn our creative process into a product that gets sold back to us.
Fix the software stack.
Respect the people who actually make records.
Stop sacrificing professional trust for speculative AI positioning.
The future of audio technology does not have to be anti-human. It does not have to be a race toward synthetic everything. It does not have to treat musicians as obstacles, engineers as legacy users, or creativity as an inefficient behavior waiting to be optimized out of existence.
The best tools have always extended human intention. They help us hear better, move faster, make sharper decisions, solve problems, and get closer to the thing we are trying to make. That is the future I want to support.
