// Yapay Zeka

Listeners don’t want AI music, but blind tests show the opposite

AI-generated music is flooding streaming platforms, while listener surveys are negative. Blind listening tests reveal that the reaction is driven not by how the music sounds, but by concerns about trust and copyright.

AI-generated music has moved beyond being an experiment and become part of the music industry. Full songs can be generated from text in seconds, and thousands of AI tracks are uploaded to streaming platforms every day. Yet listener surveys paint a distinctly negative picture. New data and blind listening tests show that this negativity is aimed not at how the music sounds, but at what it means.

The AI music explosion on streaming platforms

Deezer, which transparently labels AI-generated tracks, announced that it receives about 75,000 AI-generated tracks per day. That number corresponds to roughly 44 percent of the daily music uploaded to the platform. But more production does not mean more listening: most AI tracks get either no plays or very few.

Why listener attitudes are hardening

A study conducted between May and November 2025 that tracked attitudes toward the use of AI in music found that listener comfort declined from -13% to -20% on a net scale. Listeners are less hung up on how a song sounds than on three issues: originality, the connection with the artist, and the future of musical labor. Knowing that a real person’s emotion and story lie behind a song makes the track more meaningful; in a fully AI-generated song, that connection weakens.

The labeling effect and job security concerns

Knowing that a track is AI-generated changes perception. In an experiment using the same audio recording, participants found the track significantly more likeable, emotionally more positive, more impressive, and higher quality when they were told the music was played by a human. Listeners also make distinctions based on how AI is used: while mixing, mastering, sound design, or lyric-writing support is accepted, reactions harden toward almost entirely AI-generated songs. One reason for the concern is musicians’ job security. If companies opt for fast and cheap production, singers, songwriters, and producers may have fewer opportunities to find work.

What blind listening tests show

A survey of nine thousand people found that 97 percent of participants could not distinguish fully AI-generated music from human-made music. In another study, participants said before listening that they preferred human-made music, but after listening to both types of tracks they gave similar scores; even so, they were less willing to approve of AI tracks and recommend them to others. So the problem is not the sound, but what the music represents: trust, transparency, and the place of human labor.

The copyright and visibility debate

Copyright is a separate problem area. AI systems are largely trained on existing recordings; there is debate over whether artists’ recordings, compositions, and voices are used without permission and how contributing rights holders will be compensated. Another risk is narrowing visibility on streaming platforms. Because large numbers of AI tracks can be uploaded quickly, it may become harder for human artists to stand out and for quality work to be noticed.

The search for balance

For the industry, the sensible path is not to replace musicians with AI but to support them in the production process. In tasks such as audio cleanup, mastering, sound design, vocal processing, and restoration of old recordings, these tools save time, while creative decisions remain with humans. The listener’s clearest demand is transparency: clear labeling of whether a track is human-made, AI-generated, or a joint production of both.

Source: Unite.AI

// JOIN THE CONVERSATION

Share your experience

Bir sorunun, farklı bir çözümün veya eklemek istediğin bir notun var mı? Giriş yapmadan da yazabilirsin.

// KEEP READING

One more note.

All posts ↗︎