In a modern twist on digital fraud, North Carolina musician Michael Smith managed to scam music streaming platforms out of an estimated $12 million by leveraging artificial intelligence (AI) and bots. Between 2017 and 2024, Smith and his co-conspirators ran a complex operation involving thousands of AI-generated songs and bots to exploit streaming platforms for royalties. The intricate scheme eventually caught the attention of federal authorities, leading to charges against Smith for multiple criminal offenses.
Smith’s fraud was rooted in the massive growth of music streaming platforms, which pay artists a small fee per stream. Realizing the potential to manipulate this system, Smith, alongside his associates, devised a plan to generate royalties by using automated bots to repeatedly stream songs. The idea was simple: upload a track, have bots play it repeatedly, and collect the payouts. However, Smith knew that if he relied on just a few songs, streaming them thousands of times would likely raise suspicion with the platforms.
To avoid detection, Smith took his fraud to another level by dramatically increasing the scale of his operations. Instead of relying on a few tracks, Smith acquired hundreds of thousands of AI-generated songs. These tracks came from a co-conspirator, who was also the CEO of an AI music company. With this massive library of content, Smith could assign his bots to stream different songs in smaller quantities, avoiding anti-fraud mechanisms set by streaming services.
AI-Generated Music and Streaming Bots
Smith’s scheme centered around the use of AI-generated music. AI tools are capable of creating music tracks in a fraction of the time it takes a human artist, enabling Smith to build a vast collection of songs without the need for traditional artists or studios. This approach made it easier for him to diversify the content his bots were streaming, thereby reducing the likelihood of detection by streaming platforms’ fraud prevention systems.
To support the operation, Smith employed a network of over 1,000 bots, hosted on 52 cloud services, and routed through VPNs to mask their true locations. These bots were able to stream more than half a million songs every day. The streams were small enough per song to avoid raising red flags, but over time, the sheer volume of content and streams added up to enormous payouts.
Each stream on platforms like Spotify, Apple Music, or YouTube pays a fraction of a cent. Smith calculated that he could earn around $100,000 per month using this method. His strategy was laid out clearly in emails that were uncovered by federal authorities, where he explained that the key to avoiding detection was to generate a huge amount of content and limit the number of streams per song. “We need to get a TON of songs fast to make this work around the anti-fraud policies these guys are all using now,” Smith wrote to his associates.
Operating in Secret for Seven Years
From 2017 until 2024, Smith’s fraudulent operation went largely unnoticed. By spreading streams across hundreds of thousands of songs under fake artist names, he successfully manipulated streaming services into paying out millions in royalties. Court documents revealed that Michael Smith even boasted in February 2024 about having achieved “over 4 billion streams and $12 million in royalties since 2019.”
Smith’s fraudulent network of bots and AI music flew under the radar for years because of its strategic design. The bots, instead of focusing on a few songs, spread their attention across thousands of tracks, each streamed a small number of times. This strategy made the streams look organic, minimizing the risk of platforms flagging them for unusual activity. Each of the tracks had a unique fake artist name and title, further obscuring the true nature of the operation.
Exploiting Loopholes in the Streaming Economy
The rise of music streaming services over the past decade created opportunities for fraudsters to exploit the industry’s royalty payment systems. Smith’s scheme highlights how weaknesses in anti-fraud measures can be taken advantage of at scale. Streaming services typically use algorithms to detect fraudulent streams, but Smith’s approach of spreading streams across a vast library of AI-generated songs made it difficult for these systems to detect unusual behavior.
The economics of streaming royalties also played a role in Smith’s success. With each stream paying out only a fraction of a cent, the scheme relied on sheer volume to generate significant revenue. By using bots to play his music billions of times, Michael Smith turned what would otherwise be an insignificant payout per stream into millions of dollars over several years.
Despite the complexity of the operation, federal authorities eventually caught up with Michael Smith and his co-conspirators. The scheme’s unraveling began when suspicious patterns of streaming activity were noticed and investigated. Court documents show that the operation was well-documented in emails between Michael Smith and his associates, which laid out their tactics and the scale of their fraud.
In addition to the emails, authorities uncovered details about the bot network Smith used to carry out the operation. The bots, hosted on cloud services and protected by VPNs, were key to sustaining the fraud. With this infrastructure, Smith was able to keep his scheme running for years without detection.
A Multi-Million Dollar Fraud
By the time authorities shut down Smith’s operation, he had collected over $12 million in fraudulent royalties. The charges against him include multiple counts of fraud and conspiracy, reflecting the scale and sophistication of the scheme. The case highlights the growing threat of AI-generated content and bot networks in the digital economy, and the challenges faced by streaming platforms in detecting and preventing such fraud.
Smith’s operation serves as a warning to both streaming services and law enforcement about the potential for future fraud using AI technology. While Smith’s scheme has been exposed, the underlying issues that allowed it to succeed remain a concern for the industry going forward.
