First major US streaming fraud case ends in guilty plea after using AI to gain $8 million in fraudulent royalties
Streaming fraud is more relevant than ever in today’s AI age, and the first US case is a reminder of the problem.
Streaming fraud has long been a problem for the music industry. But now, it’s possible on a much larger scale thanks to AI. In a landmark moment for the industry, the problem is being handled in a US courtroom for the first time.
North Carolina-based Michael Smith has pleaded guilty to conspiracy to commit wire fraud after running a large-scale artificial streaming operation powered by AI-generated tracks. The result? Over $8 million gained in fraudulent royalties, and a case that serves as a warning for the industry when it comes tackling streaming manipulation.
How did we get here?
Back in September 2024, Smith was arrested and charged with multiple offences tied to his fraudulent streaming operation. He faced felony counts for wire fraud, wire fraud conspiracy, and money laundering conspiracy. Each offence carried a potential 20-year sentence, which could have combined to 60 years. Initially, he pleaded not guilty and was released on $500,000 bail.
Fast forward to now, and Smith has pleaded guilty to one count of conspiracy to commit wire fraud. He has already agreed to forfeit the $8,091,843.64 he gained in fraudulent royalties. Sentencing is scheduled for July 29, where he faces up to five years in prison.
Inside the fraudulent streaming operation
So, how did it actually work? Smith created hundreds of thousands of AI-generated tracks to game the system. He then used automated bots to stream these tracks on repeat. At its peak, the operation was potentially pulling in over $3,000 per day, close to $100,000 per month, and over $1.2 million per year.
To stay under the radar and avoid detection, Smith used more than 1,000 accounts to stream his AI tracks. These accounts then streamed hundreds of his tracks daily, collectively generating around 660,000 streams a day.
Over time, that added up to billions of fake steams, and more than $8 million in fraudulent royalties.
“Michael Smith generated thousands of fake songs using artificial intelligence and then streamed those fake songs billions of times. Although the songs and listeners were fake, the millions of dollars Smith stole was real. Millions of dollars in royalties that Smith diverted from real, deserving artists and rights holders. Smith’s brazen scheme is over, as he stands convicted of a federal crime for his AI-assisted fraud.”
Jay Clayton, a US attorney for the Southern District of New York
Why this matters for the industry
While this is the first major case of its kind in the US, it’s not the only example globally. In 2024, a case in Denmark also saw an individual jailed for artificially inflating streaming numbers. This case shows it’s not a one-off, it’s an ongoing problem.
At its core, streaming fraud directly impacts how money flows through the industry. Every fraudulent stream diverts money away from the royalty pool, reducing what real artists and songwriters earn from genuine listeners.
The bigger picture
What’s changing now is the technology. AI makes it faster and easier to generate content which, when paired with automated bots, makes streaming fraud easier than ever before. But, AI-generated tracks aren’t going anywhere, and are instead becoming a bigger part of the streaming ecosystem. Just this week, Sony announced it had removed over 135,000 AI deepfake songs of its artists from streaming platforms.
Alongside the creative opportunities that AI can bring, there’s also the risk. Earlier this year, Deezer revealed that a huge 85% of streams tied to AI-generated tracks were fraudulent in 2025. While Deezer has tools in place to detect fraudulent AI tracks and remove them from royalty calculations and limit their visibility on the platform, the same can’t be said for every platform.
Moving forward, more needs to be done across the industry. Because in a market that’s already crowded, the last thing real artists need is to compete with machines gaming the system.