I’m using Udio; there’s also Suno, but it leans more towards modern Gen Z music. I believe AI music is already capable of sounding incredibly high quality, but the services distributing these AI tools seem to be deliberately downgrading their potential (to avoid copyright complications). This is just my gut feeling, based on using Udio since its launch. Initially, the tool sounded much more creative and consistent. Now, I’m noticing that after 2 minutes, generated tracks start to exhibit issues: volume spikes, clanging bass, and overall audio degradation. The tracks become audibly harsh and screeching. Furthermore, any interesting or harmonious melodies tend to get distorted, with artificial notes and dissonance added. I suspect this is a deliberate effort to keep AI music “in its place” and prevent it from being flagged as plagiarism. However, if a service emerges that disregards copyright entirely, it could spell disaster for human musicians. But for now, of course, live musicians have nothing to worry about. The services are still limited in their capabilities and definitely don’t surpass real people.
