■ SAFE RISK ■ Creative Arts
The concert-hall composer, no; the composer of thirty-second background cues, yes, and it's already happening. AI generates serviceable functional music instantly, which splits the profession between commodity sound and authored art.
“AI composes symphonies, but nobody gives a standing ovation to a GPU cluster.”
Our AI replacement risk score — how we score jobs
Composers today mostly don't resemble the powdered-wig image. The working majority write to brief: film and TV scores against picture and deadline, game music with interactive layers, advertising cues, library (production) music sold for licensing, and arrangements. The craft involves sketching themes, orchestrating, producing convincing mockups in a DAW with sample libraries — clients approve mockups, not scores — revising to notes from directors, and preparing parts for live sessions when budgets allow. Concert composers layer commissions, grants, teaching, and premieres into a patchwork living.
Generative music has crossed the usefulness threshold. Text-to-music tools produce broadcast-passable cues in seconds, and the production-music library market — a real income floor for many working composers — is directly exposed, along with royalty pools now diluted by AI-generated tracks flooding streaming platforms. Inside legitimate workflows, AI assists rather than replaces: generating variations, suggesting harmonizations, cleaning up orchestration, speeding the mockup treadmill. Copyright remains the industry's live fault line, since purely AI-generated works sit in contested legal territory and clients need clean chains of title.
The defensible ground is authorship and collaboration. A film score is not audio wallpaper; it's a creative argument with a director conducted over months of revisions, references, and trust — the composer is hired for taste and partnership, not note production. Concert music runs on reputation, premieres, and the audience's interest in a human voice saying something. And live performance keeps its pull: ensembles commission people. Our risk score splits the difference between a commodity segment being automated now and an authored segment that AI mostly accelerates.
Automatability: our editorial assessment of current and near-term AI capability
The commodity end is being automated as we speak — library music, background cues, and low-budget scoring feel it through the late 2020s, with streaming royalty dilution compounding the squeeze. Mid-tier media scoring hybridizes: AI-assisted, human-directed. By the mid-2030s expect clear market bifurcation — machine-generated functional music everywhere, human composers concentrated where authorship, collaboration, and live performance carry the value, a structure stable into ~2040.
Yes — functional music, convincingly. Text-to-music systems generate broadcast-usable cues, ambient tracks, and genre pastiches in seconds, which is why production-music libraries and background scoring feel the pressure first. What AI doesn't do is sustain a creative collaboration with a director over months, or write from a personal voice audiences seek out. Competence is automated; authorship isn't.
The volume trades: library and production music, ad cues, royalty-free tracks, podcast and corporate scoring — anywhere the client wants appropriate sound rather than an author. Streaming royalty pools are also being diluted by AI-generated uploads. Film, game, and concert work with meaningful creative collaboration is pressured but far more resilient, because the relationship is the product.
Use them, selectively. Working composers already lean on AI for variations, orchestration suggestions, and faster mockups — the mockup treadmill was the profession's least-loved labor anyway. The strategic line to hold is authorship: keep your creative decisions documented and your output clearly human-directed, since clean provenance increasingly matters for both copyright and client trust.
Viable, with the ladder's bottom rungs sawed off. The traditional entry path — grinding out library tracks and small cues — is exactly what generative tools automate, so emerging composers must reach collaboration-based work faster: short films, indie games, ensembles, direct commissions. The patchwork career (media work, commissions, teaching, performance) remains the realistic model; pure cue-writing no longer sustains it.