Artificial intelligence is now producing music faster than audiences can listen to it, with fully AI-generated tracks accounting for more than half of all new music delivered to streaming platform Deezer on peak days.
Deezer received an average of almost 90,000 completely AI-generated songs each day during June 2026, up from around 75,000 a day and 44% of uploads in April.
The figures mark the first time synthetic music has surpassed human-made releases among the platform’s incoming tracks.
The rapid increase highlights how generative music tools have removed many of the traditional barriers to producing and distributing a song. Users can now create vocals, lyrics and instrumentals from a simple prompt before uploading the finished track to a streaming service.
However, the explosion in supply does not necessarily reflect genuine listener demand.
Streaming fraud clouds AI boom
Deezer previously found that as much as 85% of streams involving fully AI-generated music were fraudulent, often involving automated accounts designed to inflate plays and divert royalty payments. The platform said synthetic tracks represented only a small proportion of genuine listening despite their growing share of uploads.
This creates a potential problem for streaming services that distribute subscription revenue through shared royalty pools. Even relatively small amounts of manipulated activity can direct money away from legitimate musicians and songwriters.
Deezer has responded by detecting and tagging fully AI-generated tracks, excluding them from editorial and algorithmic recommendations and removing fraudulent streams from royalty calculations.
The company is also moving to delete synthetic tracks associated with manipulation, along with AI-generated uploads that receive no streams for six months.
Its approach suggests platforms may not need to ban artificial music outright, but will increasingly separate content that listeners actively choose from material uploaded primarily to exploit payment systems.
Human-made music under pressure
The issue extends beyond streaming fraud.
Musicians and rights holders argue that many generative music systems were developed using copyrighted recordings without permission or payment. Once trained, those systems can produce songs that imitate existing genres, performers and production styles at enormous scale.
That debate has become particularly relevant in Australia, where artists have called for stronger protections over the use of their work in AI training datasets.
The Federal Government has pledged to prevent technology companies from using Australian books, music, art and journalism to train AI systems without authorisation, while plans for a new Office of AI are expected to place copyright and creator compensation at the centre of future regulation.
For musicians, the concern is not simply that an AI song could become a hit. It is that unlimited low-cost content could make it harder for emerging artists to be discovered, while simultaneously weakening the royalty system on which their careers depend.
Platforms face an authenticity test
Streaming companies will now need to decide how synthetic music should be identified, recommended and monetised.
Clear labelling may allow listeners to choose between human and AI-generated work, while better detection systems could help prevent fake streams. More difficult questions remain over training data, attribution and whether artists should receive compensation when their work influences an AI-generated track.
The milestone recorded by Deezer is unlikely to be the peak. As music-generation tools become cheaper and more accessible, synthetic uploads could continue to grow far faster than human releases.
The challenge will be ensuring that technological creativity does not overwhelm the people whose music helped make the technology possible.