Artificial Streaming, Not AI, Is the Immediate Threat to Music’s Integrity (Guest Column)

Last month, Reddit detectives made a startling discovery, and in so doing perhaps offered a glimpse at the future of music: an AI-generated band, The Velvet Sundown, was landing in users’ eardrums. With a full album of decently-polished tracks, a made-up bio, and plausibly real images, “they” had quickly racked up 1  million monthly listeners on Spotify, seemingly out of thin air. 

As the industry raced to get to the bottom of The Velvet Sundown’s success, questions were raised and debate ensued. The industry and the internet easily found similar bands and artists, who, with nary a real world footprint, had racked up hundreds of thousands of listeners across all platforms and genres.

The discussion quickly zeroed in on questions about human creativity, IP protections, ethics and the role of AI in music. But the immediate problem, the one actually costing artists money, isn’t how these songs were made. It’s how they were heard. 

The problem isn’t artificial intelligence. It’s artificial streaming. 

Whether music is created by machine or human, inauthentic streams distort engagement and siphon royalties. Bots, clickfarms and manipulated playlists are common tactics being used across the digital ecosystem to manufacture popularity. By some estimates, 5 to 10 percent of all streams are inauthentic. That’s billions of listens, and real royalties being redirected from legitimate creators to fraudsters jumping into the royalty pool. It’s organized and sophisticated manipulation that’s leading to lawsuits and criminal investigations. 

Don’t get me wrong: As U.S. authorities allege in the Michael Smith case, AI-generated music exacerbates the ease of committing fraud by pulverizing any barriers to creating music. But ask any artist, and they’ll tell you that the mere creation of a song doesn’t guarantee any listeners, let alone hundreds of thousands of listeners. In a market flooded with hundreds of thousands of new tracks each day, streams don’t happen by magic. 

If we want to have a truly fair and transparent marketplace, we must scrutinize whether any artist or any track benefits from inauthentic streams. We need to ensure that a song, regardless of its creation, earns listeners through genuine engagement, not gamesmanship. Because, ultimately, fraudsters don’t care if the music is made by a robot or a rock band. They only care about exploiting the system. 

Fortunately, this is one area where the industry is able to actively take charge and fight back. Companies across the ecosystem are investing heavily in fraud detection tools and technologies. Streaming platforms are penalizing fraudulent tracks to deter bad actors and working with artists, labels and distributors to raise awareness about suspicious activity. And through the Music Fights Fraud Alliance, industry leaders are collaborating on anti-fraud solutions with a goal to disrupt fraud as a united front. 

Through collaboration and shoring up our defenses, the music industry can root out bad actors and deter their tactics. Together, we can fight fraud and secure a more fair, transparent, and authentic streaming future. 

Michael Lewan is the Executive Director of the Music Fights Fraud Alliance. Launched in 2023, the MFFA is a global non-profit organization representing more than 25 member companies aligned in a mission to eradicate streaming fraud. Prior to joining the MFFA, he worked at the Recording Academy from 2017 to 2025 where he led public policy and advocacy initiatives.