AMERICAN MUSICIAN MICHAEL SMITH JAILED 18 MONTHS FOR USING BOTS TO STREAM AI SONGS BILLIONS OF TIMES AND POCKET $8 MILLION

American musician and business owner Michael Smith has been sentenced to 18 months in federal prison after admitting that he built a large scale music streaming fraud operation that used thousands of automated accounts to play hundreds of thousands of AI generated songs billions of times, generating more than $8 million in royalty payments that prosecutors said should have gone to legitimate artists, songwriters and other rights holders.

Smith, 54, of Cornelius, North Carolina, was sentenced in Manhattan on October 6 by United States District Judge John G. Koeltl after pleading guilty in March to one count of conspiracy to commit wire fraud. In addition to the prison sentence, Smith was ordered to serve two years of supervised release and forfeit $8,091,843.64, making the case one of the most significant criminal prosecutions to emerge from the collision between artificial intelligence, automated streaming and the modern music royalty system.

Federal prosecutors said Smith operated the scheme from 2017 until 2024 and gradually developed a sophisticated system capable of imitating the behavior of genuine music listeners across major streaming platforms. The operation relied on thousands of fake accounts, automated software and a rapidly expanding catalog of songs that Smith controlled, allowing him to repeatedly generate streams and collect royalty payments without a genuine audience behind the listening activity.

The platforms targeted by the scheme included Spotify, Apple Music, Amazon Music and YouTube Music. Prosecutors said Smith created fake user accounts with false email addresses and fraudulently obtained debit cards, then used software to make those accounts continuously play music he owned or controlled. At certain points in the scheme, he was operating as many as 10,000 bot accounts simultaneously.

The fraud exploited one of the basic mechanics of the music streaming economy. Every legitimate stream can generate a small royalty payment that is eventually distributed among artists, songwriters, publishers, labels and other rights holders. When automated systems create fake streams, the problem is not limited to inflated popularity figures because the false plays can also redirect money from the pool of royalties that would otherwise be distributed based on genuine listening activity.

Smith initially faced a practical problem because repeatedly streaming a relatively small number of songs millions of times could attract attention from anti fraud systems. Prosecutors said he understood that concentrating large numbers of plays on a limited catalog would make the manipulation easier to detect, so he sought what he described in an email as a large quantity of songs that could help him avoid the streaming platforms anti fraud policies.

Artificial intelligence became central to solving that problem. Smith worked with a co conspirator who supplied him with hundreds of thousands of AI generated tracks that could be uploaded to streaming services. The songs could then be distributed across a large catalog, allowing the bot accounts to spread their activity over many tracks rather than repeatedly streaming the same limited set of recordings.

The songs were also given invented artist names and track titles so that they appeared to come from real musicians rather than from a mass production system built around artificial intelligence. Court documents identified examples of the fake artists as Calliope Bloom, Calm Baseball, Calm Force, Calvin Mann, Calypso Xored, Camaxtli Minerva and Camel Edible.

Prosecutors also disclosed examples of the unusual song titles attached to the AI generated recordings, including Zygophyceae, Zygophyllaceae, Zygophyllum, Zygotenes, Zygotes, Zygotic, Zygotic Lanie, Zyme Bedewing, Zymopure and Zyzomys. The names were randomly generated and attached to audio files as part of an effort to make the catalog look like the work of a large number of independent artists rather than one coordinated streaming operation.

The underlying AI files were often initially identified only by strings of random letters and numbers. Smith then supplied metadata, including artist names and song titles, so the recordings could be presented on streaming platforms as ordinary releases. Prosecutors said the strategy allowed him to create the appearance of a sprawling music catalog while maintaining control over the streams generated by his automated accounts.

Court filings show that one of Smiths collaborators described the operation in unusually candid terms. In a 2019 email, the collaborator told Smith that what they were producing was not conventional music but instant music, reflecting the emphasis on producing large quantities of tracks quickly enough to feed the automated streaming system.

The volume of fake activity eventually became enormous. Prosecutors said Smiths AI generated songs were streamed billions of times by the bot accounts he controlled, producing royalty payments that accumulated over several years. At one stage, the system was capable of generating hundreds of thousands of artificial streams every day.

The scale became particularly striking in April 2023. According to the Justice Department, Taylor Swifts entire catalog received approximately 9.3 million YouTube Music streams through family plan accounts that month. During the same period, Smiths bot network generated about 80.9 million family plan streams of his AI generated music, nearly nine times the figure attributed to one of the worlds most commercially successful recording artists.

Those numbers did not represent a sudden global audience for Smiths catalog. Federal investigators said the overwhelming majority of the listening activity was produced by automated accounts programmed to repeatedly stream the music and create the appearance of genuine consumer demand.

Smith also took steps designed to hide the manipulation. Prosecutors said he spread streams across thousands of songs so no single track would show an obviously abnormal level of activity. He also used false identities and payment information when creating streaming accounts and represented those accounts as genuine subscribers even though they were being controlled by automated software.

The fraud therefore depended on both artificial content and artificial listeners. AI technology provided the enormous catalog of songs, while bots supplied the audience. Together, the two systems created a closed commercial loop in which machine generated music was being played by machine controlled accounts and converted into real royalty payments.

The Justice Department has emphasized that the case was not brought simply because Smith used artificial intelligence to produce music. AI generated music is not automatically illegal, and musicians and producers increasingly use artificial intelligence for songwriting, production, vocals, mastering and other creative processes. Smiths criminal liability arose from the deliberate manipulation of streaming platforms and the use of deceptive accounts to generate fraudulent royalty payments.

That distinction has become increasingly important as AI generated music spreads across digital platforms. Streaming services are receiving growing volumes of computer generated material, while labels, artists and regulators are debating questions involving copyright, royalties, disclosure and the economic impact of automated content.

Smiths case stands apart because prosecutors said the technology was intentionally combined with fraud. His bot accounts were designed to mimic legitimate users, the artist identities were fabricated, the streams were artificial and the resulting royalty payments were generated through activity that did not reflect genuine consumer listening.

The case first became public in September 2024 when Smith was charged by federal prosecutors in the Southern District of New York. Authorities described the case at the time as the first criminal prosecution centered on artificially inflated music streaming aided by artificial intelligence.

The original indictment included allegations of wire fraud conspiracy, wire fraud and money laundering conspiracy. Prosecutors initially said the broader scheme had generated more than $10 million in fraudulent royalty payments over the years it operated.

Smith ultimately pleaded guilty on March 19, 2026 to one count of conspiracy to commit wire fraud. Under the plea arrangement, he accepted responsibility for obtaining more than $8 million through the scheme and agreed to the forfeiture amount later imposed by the court.

That distinction is important when reporting the financial scale of the case. Prosecutors initially alleged that the broader operation generated more than $10 million, but the amount formally tied to Smiths guilty plea and sentencing was $8,091,843.64.

During sentencing, Smith became emotional as he addressed the court and acknowledged the damage caused by his actions. He described himself as someone who had wanted to become a successful musician, husband and father while earning the respect of the music community, but admitted that his conduct had betrayed those ambitions.

Smith said he had failed in his responsibilities and acknowledged that he had transformed his love of music into what he described as a perverse business model. His lawyers asked the court to consider a sentence without imprisonment, arguing that the regulatory environment surrounding artificial streaming had been less clearly defined when Smith began the operation and pointing to his acceptance of responsibility.

Federal prosecutors sought a significantly longer prison term of 46 months. They argued that incarceration was necessary not only to punish Smith but also to discourage others from exploiting streaming platforms as AI makes it easier to produce enormous catalogs of music at relatively low cost.

Judge Koeltl ultimately imposed an 18 month sentence, substantially below the prison term sought by prosecutors but rejecting the defense request for probation. Smith will also spend two years under federal supervision following his release.

The sentence comes as the music industry faces growing concern over streaming manipulation. Fraudulent streams can distort charts, recommendation systems, popularity metrics and royalty distributions, creating economic consequences for legitimate artists whose music is competing for attention and payment against artificially inflated content.

The rapid improvement of generative AI has intensified those concerns because it is now possible to create large quantities of music far faster and more cheaply than traditional recording methods allow. While that technology can support legitimate creativity, the Smith case demonstrates how the same tools can be combined with automation to exploit systems designed around genuine human listening.

Streaming companies have responded by increasing fraud detection efforts and tightening rules governing artificial activity. Platforms routinely remove suspicious streams, withhold royalties and suspend accounts linked to manipulation, but the Smith case shows that large operations can still generate substantial sums before being identified.

The prosecution also raises questions about how streaming platforms verify identities, payment methods and listening behavior at scale. Smith was able to maintain thousands of fake accounts and distribute streams across a massive catalog, suggesting that fraud detection systems must increasingly analyze networks of activity rather than focusing only on unusually popular individual songs.

For artists, the case is particularly significant because streaming royalties are distributed within an ecosystem where every fraudulent play can affect how money is allocated. Artificial streams effectively compete with genuine listeners for portions of the royalty pool, meaning large scale manipulation can reduce the revenue available to legitimate rights holders even when no single artist can identify an exact financial loss.

The fake artist names and bizarre song titles revealed in the court documents also illustrate how little conventional artistic identity was required for the scheme to operate. Calliope Bloom, Calm Baseball, Calvin Mann and the other invented artists did not need fan bases, tours, publicity campaigns or genuine listeners. They existed primarily as labels attached to files that bots could repeatedly stream.

The same was true of songs carrying names such as Zygotes, Zygotic Lanie, Zyme Bedewing and Zymopure. Their commercial value came not from human demand but from the ability of Smiths automated system to manufacture the appearance of demand and convert it into royalty payments.

Federal authorities have presented the case as a warning that technology does not create a legal shield for conventional forms of fraud. The methods may have involved AI generated music and automated streaming, but prosecutors treated the underlying conduct as a scheme built around false representations and the diversion of money through deceptive activity.

Smiths conviction is likely to become an important reference point as regulators and the music industry confront similar cases involving automated content and manipulated engagement. It demonstrates that prosecutors can apply existing wire fraud laws even when the technology used to carry out the scheme is relatively new.

The sentencing does not resolve the wider debate over AI music. Questions remain over ownership, copyright, training data, disclosure requirements and whether streaming platforms should treat machine generated music differently from human recordings. Those issues will continue evolving as technology becomes more capable and the volume of AI generated material increases.

What the Smith case does establish is a clearer boundary around artificial streaming. Creating or distributing AI music is one issue, but using fake accounts and bots to manufacture billions of plays and collect royalties is another. Federal prosecutors have now successfully pursued that conduct as criminal fraud.

Smith will serve 18 months in prison and forfeit more than $8 million, but the wider impact of the case is likely to extend well beyond his sentence. The operation demonstrated how generative AI, fake identities and automated listening can be combined to exploit the global streaming economy on a scale that would have been difficult to achieve only a few years earlier.

For streaming companies and the music industry, the next challenge will be preventing similar schemes before they generate millions of dollars rather than relying on criminal prosecution after the money has already been paid. As AI makes mass music production increasingly accessible, detecting whether the audience behind that music is real may become just as important as determining who or what created the songs in the first place.