'Face the music': North Carolina man accused of $10 million AI-aided streaming fraud

A North Carolina man is accused of creating "hundreds of thousands of songs with artificial intelligence" and using "bots" to stream the AI-generated tunes billions of times, federal prosecutors announced.

Michael Smith, 52, of Cornelius, North Carolina, fraudulently obtained over $10 million in royalty payments through the scheme he orchestrated from 2017 to 2024, according to a federal indictment filed in the Southern District of New York.

Smith was arrested on Wednesday and charged with wire fraud conspiracy, wire fraud and money laundering conspiracy, a Justice Department news release said. Each offense carries a maximum sentence of 20 years in prison.

“As alleged, Michael Smith fraudulently streamed songs created with artificial intelligence billions of times in order to steal royalties," U.S. Attorney Damian Williams said in the release. "Through his brazen fraud scheme, Smith stole millions in royalties that should have been paid to musicians, songwriters, and other rights holders whose songs were legitimately streamed. Today, thanks to the work of the FBI and the career prosecutors of this Office, it’s time for Smith to face the music.”

Smith did not have a defense attorney listed in court records.

Target thefts:19 adults, 3 teens accused in massive retail-theft ring at Target stores

How did Michael Smith execute the scheme?

To carry out the scheme, Smith created thousands of "bot accounts" on music streaming platforms — including Amazon Music, Apple Music, Spotify, and YouTube Music, according to the indictment. He then used software to make the accounts constantly stream the songs he owned, the court document says.

Smith estimated that at one point he could use the accounts to generate about 661,440 streams per day, yielding $1,207,128 in annual royalties, according to the Justice Department release.

To avoid the streaming of a single song, Smith spread his automated streams across thousands of songs, the indictment says. He was mindful that if a single song were to be streamed one billion times then it would raise suspicions among the streaming platforms and music distribution companies, the court document continued.

A billion fraudulent streams spread throughout tens of thousands of songs would be more difficult to detect due to each song being streamed a smaller amount of times, prosecutors said. Smith soon identified a need for more songs to help him remain under the radar, according to the Justice Department.

On or about December 26, 2018, prosecutors said Smith emailed two coconspirators, writing “We need to get a TON of songs fast to make this work around the anti-fraud policies these guys are all using now."

Prosecutors: Michael Smith turned to AI to keep the scheme afloat

To ensure Smith had the necessary number of songs he needed, he eventually turned to AI. In 2018, he began working with a chief executive officer of an AI music company and a music promoter to create hundreds of thousands of songs using artificial intelligence that he could then fraudulently stream, according to the indictment.

The promoter would provide Smith with thousands of songs each week that he could upload to the streaming platforms and manipulate the streams, the charging document says. In a 2019 email to Smith, the promoter wrote: “Keep in mind what we’re doing musically here… this is not ‘music,’ it’s ‘instant music’ ;).”

Using the hundreds of thousands of AI-generated songs from the promoter, Smith created randomly generated song and artist names for audio files so it would seem as if the music was created by real artists, according to the indictment.

Some of the AI-generated artist names included “Calliope Bloom,” “Calliope Erratum,” “Callous,” “Callous Humane,” “Callous Post,” “Callousness,” “Calm Baseball,” “Calm Connected,” “Calm Force,” “Calm Identity,” “Calm Innovation” and “Calm Knuckles,” the U.S. Attorney's Office said.

Smith would lie to streaming platforms during the scheme, including using fake names and other information to create bot accounts and agreeing to abide by terms and conditions that prohibited streaming manipulation, the Justice Department said. He also caused the streaming platforms to falsely report billions of streams of his music, while in reality, he knew the streams were from his bot accounts as opposed to real human listeners, according to prosecutors.