AI-Driven Streaming Fraud: Guilty Plea in $10M Royalty Heist

The guilty plea of Michael Smith, a 52-year-old North Carolina resident, represents a watershed moment for the intersection of generative AI and digital intellectual property law. By leveraging AI to generate hundreds of thousands of songs and deploying a sophisticated network of bot accounts to stream them billions of times, Smith managed to extract more than $10 million in royalty payments from giants like Spotify, Apple Music, and Amazon Music. This case is not merely a story of simple theft; it is a demonstration of how generative AI can be weaponized to exploit the algorithmic foundations of the modern creator economy.

At the heart of the scheme was a technical workaround designed to bypass the 'content fingerprinting' technology used by streaming services to detect repetitive or fraudulent uploads. Smith realized that if he uploaded a small number of songs and streamed them millions of times, the platforms would quickly flag the activity as suspicious. To counter this, he utilized AI tools to churn out a massive volume of distinct, short tracks—often little more than computer-generated noise or simple melodies. By spreading billions of streams across hundreds of thousands of unique AI-generated songs, he kept the play counts for any single track low enough to evade automated fraud detection systems while still accumulating massive aggregate royalties.

“

Prosecutors revealed that at the height of the scheme, Smith was generating approximately 661,000 streams per day, yielding an estimated $3,371 in daily royalties, or roughly $1.2 million per year.

The scale of the operation was industrial. Prosecutors revealed that at the height of the scheme, Smith was generating approximately 661,000 streams per day, yielding an estimated $3,371 in daily royalties, or roughly $1.2 million per year. To facilitate this, he managed a fleet of over 10,000 bot accounts, using software to automate the playback process across multiple computers and VPNs to mask their geographic origins. This 'streaming farm' approach effectively diluted the royalty pool for legitimate artists, as streaming platforms typically distribute a fixed percentage of their total revenue based on the share of total streams an artist receives.