AI can generate sound, but only a person can decide what the work should express — remember that music has survived this before; the listener still decides.