How will young people learn music and production in a post-AI music industry?“I’m excited for AI that can cut out the annoying tasks that get in the way of the creative process, and I also can see a world where it becomes a sort of writing partner. But, in regards to game development, I think it may cut out many opportunities for newer composers.”
I’m talking to Michael James Collier, a composer at Grinding Gear Games – the studio behind the popular action-RPG game, Path of Exile.

READ MORE: Will RIAA’s lawsuit against Udio and Suno really be the win we’re hoping for?

The sentiment he’s expressing is common. I’ve heard versions of it from many people across the music industry: AI has amazing potential to streamline and automate away the boring drudgery, but may simultaneously wipe out anyone without sufficient skill to justify their pay check.
At the same time, there also seems to be a broad consensus that, no matter how convincing AI becomes, it will never replace the true masters. Even if AI achieves intellectual superiority, we will still want Hans Zimmer to score the new blockbuster, Rick Rubin to produce the hit of the summer and Bob Ludwig to master it. Premium projects will always require a human touch.
But the Steve Albini’s of the world do not arrive fully formed. Skills have to be learned, talents need to be honed, and, historically, this has been done through on-the-job training. Composers work their way up from scoring student films, to trailer music, to commercials, and, finally, to well-paid projects in film, TV, and gaming. Mix engineers might spend their days editing and cleaning up drum tracks while interning at a studio, or training their ear mixing and mastering the DIY recordings of local garage bands.
A photo from 2005 of Steve Albini in his Electrical Audio studio. Image: Paul Natkin/Getty Images
They do. Or, at least, they were.
These ‘lower-tier’ tasks are first in the firing line for AI automation. So, we need to stop and ask ourselves: if the bottom ten rungs of the ladder are broken, then exactly how will the next generation of music profession
These initial projects don’t offer much money or prestige, but they are foundational to how music professionals build a portfolio, establish industry connections, and, crucially, get good at what als achieve the kind of audio mastery that none of us want to see go extinct?
One scenario is that we simply redefine what counts as low-level. Using the latest AI tools, rookies may be able to take up tasks that would have once required a seasoned pro. In turn, skilled professionals will be freed up to produce radically new kinds of material, defining next-gen workflows as they do so.
For historical parallels, look to the Arts and Crafts movement of the 19th century. Reacting to the explosion of mass produced goods, some manufacturers opted to integrate the tools of industrialisation with traditional craftsmanship. That practice not only produced wholly new styles of design and art, it also kept those craftspeople relevant in the new economy; empowering them to retain a measure of control over their work.
It’s not impossible to see a similarly positive outcome happening in the world of audio as young people increasingly explore AI for its creative possibilities. Futuristic DAW’s like RipX genuinely break new ground when it comes to what we can do with existing audio, audio restoration plugins remove laborious tasks that no one enjoys, and a dizzying array of new production tools are reshaping how we write, record, and refine music. These innovations have clear potential to supercharge human creativity rather than supplant it.
At the same time, we cannot ignore the potential AI clearly has to eviscerate whole sections of the creative industries, and, if we once again cast an eye back to the history of industrialisation, then a far darker scenario emerges.
The Luddites – much maligned as technophobes who wanted to resist any innovation – were actually skilled workers who, over a relatively short period of time, saw their lives fundamentally reshaped from that of skilled artisans, to easily replaceable menials who performed repetitive tasks for drastically lower pay.
Which brings us to AI song generators.
Image: DisobeyArt via Getty Images
Leaving aside the ethical can of worms that powers text-to-audio song generators, it’s worth noting that they exist in wholly different categories to things like AI stem separation. Why? Because, despite what their fundraising announcements might say, their intended users are not musicians.
Companies like Suno and Udio may highlight their many individual users, but as the initial novelty wears off, it’s very hard to see them sticking to their current subscription model. Hell, Spotify boasts almost half a billion loyal subscribers and still runs at a loss year after year. If they are not sued out of existence, or if they manage to cut a licensing deal with the majors, then, sooner or later, they will turn to where the real money is: enterprise customers who need large amounts of music but don’t want to pay humans to write or record it.
You can’t retrain for an industry that is actively working to uncouple you from the creative chain. More to the point, for music professionals to retrain as prompt engineers would be the exact opposite of up-skilling, it would be de-skilling.
Within a few years, it may be possible to produce much of the background music that we hear in everyday life – in commercials, supermarkets, or company training videos – through de-skilled prompt engineering. But that world would leave frighteningly few opportunities for people to gain the nuts and bolts experiences needed to become the next John Williams.
If society is to keep pace with the AI industry’s ‘move fast and break things’ mantra, then thorny questions around training data, copyright, labour projections, and the value of human artistry will all need to be addressed sooner rather than later.
One thing is clear: you cannot remove the grass-roots level without destabilising the entire ecosystem. If AI shatters our established methods of learning and career building, then there is no blueprint for how we put those pieces back together.
The post How will young people learn music and production in a post-AI music industry? appeared first on MusicTech.