Fields / City
First stop
Music is the field that takes something interior — a feeling, an idea, a tension, a mood that has no words — and gives it a form other people can hear and feel. A song, a symphony, a film score, a beat, a worship set, a lullaby: each is an inner state made audible and handed to someone else.
That is the structural pull at the field level, and it is why the primary gradient is Expression. Whatever a particular music job looks like day to day, the field exists because human beings need to externalise what is inside them and to receive what is inside others, and music is one of the oldest and most direct technologies for doing it. The move from internal to expressed is the through-line that runs under the performer on a stage, the composer at a desk, the producer at a console, the teacher in a practice room, and the therapist in a hospital ward alike.
Creation and Aesthetic sit immediately alongside Expression as the field's other defining pulls. A great deal of music work is bringing something genuinely new into existence — a melody that did not exist yesterday, an arrangement assembled from nothing — which is the Creation gradient, and a great deal of it is the patient pursuit of beauty: refining a phrase, balancing a mix, shaping a tone until it is exactly right, which is the Aesthetic gradient. Connection runs underneath all of it, because music is rarely a solitary destination. It is made with others and made for others, and the felt purpose of most of the work is to bridge the distance between people — to make a room of strangers feel one thing at the same moment. The honest framing of the field holds all four: music is where expressing, creating, beautifying, and connecting meet in the same act.
The daily texture is far more repetitive, technical, and businesslike than the romance of the field suggests. The visible moment — the performance, the release, the finished track — sits on top of enormous quantities of invisible, unglamorous practice and admin. A working musician spends most hours not performing but rehearsing the same passage until the hands stop thinking, maintaining technique, learning repertoire, setting up and breaking down equipment, chasing invoices, and managing the small business that a music career almost always is. The skill that separates people who sustain a life in music from people who drift out of it is as much organisational and relational as it is musical: the ones who last are the ones who treat reliability, communication, and showing up prepared as part of the craft.
What nobody tells you is that almost no one earns a living from a single music activity, and that the economics of recorded music in particular are brutal. Streaming pays fractions of a penny per play, so recordings now function mostly as marketing for the things that actually pay — live performance, teaching, session work, sync licensing, and commissions — and the typical sustainable career is a portfolio stitched together from several of these at once. Reported averages capture this only crudely: UK survey figures put the average musician or singer salary at roughly the mid-twenty-thousands of pounds, with a very wide spread from the mid-teens to the high tens of thousands depending on what mix of work someone has assembled [from salary and job-posting data, 2025]. The Musicians' Union publishes recommended minimum rates — for example a recommended minimum of around £44 an hour for individual teaching in 2025/26, and minimum engagement fees for live and session work — precisely because the open market for music labour drives rates down without them [MU 2025-26]. The students who build durable careers are usually the ones who understood early that the income is patchwork and planned for it, rather than the ones who waited to be discovered.
Music is unusual among fields in that it has no single mandatory credential for most of its roles. Nobody needs a licence to write a song, play a gig, or produce a record, and many of the most successful musicians are self-taught or conservatoire dropouts; talent, portfolio, and reputation carry more weight than any certificate. That said, formal training routes are real and matter for specific paths. In the UK and Portugal the classical and academic route runs through conservatoires and university music degrees — in Portugal a licenciatura em música at institutions such as the Escola Superior de Música de Lisboa, ESMAE, or the Universidade de Évora, often built on prior conservatório study [official]. Where music meets a regulated profession, the gate is firm: teaching music in a UK state school requires a degree plus Qualified Teacher Status [from salary and job-posting data, 2025], and practising as a music therapist requires an HCPC-approved postgraduate qualification and statutory registration before the title can be used. The realistic version of the path is that the artistic core of the field is open and merit-driven, while its institutional and clinical branches are gated like any other profession.
The pull
Internal → Expressed
The pull to give inner experience external form
Creation and Aesthetic sit immediately alongside Expression as the field's other defining pulls.
The line forks
Service update
What is newly possible in Music, what stays yours, and how AI is changing the way in. In that order, and only if you want it.
3 barriers have dropped
Professional music production historically required expensive studios, equipment, and specialist engineers
AI mastering (LANDR), mixing (iZotope), and composition tools (Suno, AIVA) reduce cost and skill requirements for professional-quality audio production to near-zero
Solo artists produce and distribute professional-quality music without label, studio, or engineer
Almost certain to keep going this way. For simple background music it has already happened, and it is still spreading to more complex production.
When making music gets cheap, far more music gets made, so getting noticed by listeners becomes the new barrier. More music does not mean more musicians can earn a living from it.
Background/commercial/library music required commissioning human composers or licensing from production music libraries
AI music generation produces functional background music from text prompts at near-zero marginal cost, eliminating the need to commission human composers for most functional music
Content creators generate custom soundtrack material without music production skills or composer commissioning
Almost certain to keep going this way. It is well documented and speeding up.
This directly takes work away from composers who wrote background and commercial music. Human composition keeps its creative and cultural value, but demand for it in this kind of work is falling sharply.
Music distribution required label or distributor relationships to reach streaming platforms
Digital distribution platforms (DistroKid, TuneCore) already weakened this; AI amplifies by combining creation + distribution and enabling anyone to generate and distribute music
Artists distribute globally and generate marketing materials without intermediaries
Almost certain to keep going this way. The old control over who gets music onto streaming services is not coming back.
Getting music out is no longer the barrier, but getting heard now is. With almost unlimited music available, listeners' attention is scarce, which favours artists with a strong identity and a real bond with their audience.
These are real jobs that exist now and did not exist before the current wave of AI. Tap one to see what it is and how well we know it.
Prompts, curates, selects, and edits AI-generated music to meet creative briefs. Human provides taste, context, quality control; AI provides raw material.
Job postings mentioning AI music supervision; agency workflows incorporating AI-generated music with human curation · early signal
Manages consent, copyright, and licensing issues around AI-generated music at labels, publishers, and platforms
Label legal teams expanding; platforms hiring for AI policy roles · early signal
Producer integrating AI tools (stem separation, AI mixing, generative composition) as core workflow elements alongside traditional production skills
Producer workflows increasingly incorporating AI tools · current fact
Bars above the line are the parts of this work that still need a person. Bars below it are what AI can already do. Tap any column to see the actual work behind it.
The honest read. Protection is sharply archetype-dependent. Live performers are protected by embodied presence. Music therapists by relational judgement and regulation. Artists with distinctive voices by irreplaceable human significance. Producers of standard commercial music have the weakest protection — their output is closest to what AI can replicate. The field's human core is real but does not protect all archetypes equally.
Music is one of the creative fields where generative AI most directly produces the core output. Text-to-music models can generate complete songs; AI mastering is already standard; AI stem separation is routine. The disruption is unevenly distributed: composition and production of commercial/background music are highly exposed, while live performance, music therapy, and the relational dimensions of teaching are structurally protected. The field's AI story is a story of asymmetric disruption by archetype.
Harder to enter than it was, and that applies fairly evenly across the ways in.
That is everything we currently know about AI in Music. It shows where things are moving so you can choose which way in suits you.
The job market
People and jobs. Generative AI is restructuring the composition/production segment. Background/commercial music composition faces near-zero marginal cost AI competition. Streaming economics under pressure from AI catalogue growth (44% of Deezer uploads AI-generated). Live performance relatively stable. Music therapy demand stable with regulatory protection. New roles emerging (AI music supervision, AI rights management). BLS projects little or no employment change for musicians through 2034.
The main limit on getting in. Market competition is the binding constraint for most music careers — intense before AI, intensified by AI. Exception: music therapy (credential gate — HCPC/CBMT) and classroom teaching (QTS).
Entry-level jobs. Contracting for commercial/functional composition and standard production work. Stable for live performance, teaching, and music therapy. Field-aggregate trajectory is contracting because the most accessible entry-level commercial work (production music, jingles, basic studio engineering) is the most AI-exposed segment.
Cost of training. Music degrees (BMus, MMus) carry significant debt relative to median musician income (BLS ~$40K, highly variable, many part-time). Portfolio career reality means income is uncertain and irregular. Debt-to-income is a serious consideration for career sustainability.
Unlike shortage fields (nursing, physiotherapy) where AI is positioned as efficiency augmentation, music has no workforce shortage. AI enters as a direct competitor producing the same output at lower cost. This makes the displacement narrative more acute and the augmentation narrative less convincing for the exposed segments.
Interchange
People drawn to Music are often drawn to these.
End of the line
That is a conversation about this field, not a test. Or keep going on your own. Both doors stay open.