The Copyright Workplace is probing how generative AI helps fraudsters scale pretend streaming schemes.
The U.S. Copyright Workplace has opened a public inquiry into music streaming fraud, asking artists, file labels, streaming platforms and anybody else with knowledge to weigh in on how bot farms and AI-generated tracks are siphoning cash from the royalty pool.
Printed within the Federal Register on October 7, the discover follows a Could 21 letter from Rep. Scott Fitzgerald of the Home Judiciary Committee. He requested the Workplace to look at how widespread streaming fraud is throughout digital platforms and what it’s doing to the trade.
The Workplace says the responses, together with its personal analysis, will assist inform Congress.
Why the pie retains getting smaller for sincere artists
The issue comes right down to how streaming pays out. Most providers put a share of their income right into a pool and divide it by play rely, so extra streams means a much bigger lower. That setup offers unhealthy actors a transparent incentive to pretend performs.
The discover lists the same old ways: bot farms, click on farms staffed by human employees, playlist stuffing, hijacked accounts and “ghost monitoring,” the place somebody uploads a barely altered copy of one other artist’s music and claims it as their very own. Many of those schemes, the Workplace notes, are run via unethical advertising corporations.
The stakes are large. The discover cites roughly $9.5 billion in U.S. streaming income for sound recordings and $22 billion globally. Music publishers, in the meantime, introduced in an estimated $7.3 billion domestically and $10 billion worldwide.
Past the royalty dilution, the Workplace factors to secondary injury. Faux streams can skew playlist placement and advice algorithms, they usually muddy the info artists and venues use to determine when and the place to tour.
AI? No manner
The inquiry leans closely on the AI angle. The discover factors to the case of Michael Smith, a North Carolina man who uploaded a whole bunch of 1000’s of AI-generated songs and used 1000’s of bot accounts to stream them and picked up about $10 million in royalties earlier than pleading responsible to conspiracy to commit wire fraud in March.
The Workplace additionally quotes the congressional letter’s view that there’s “nothing inherently mistaken” with utilizing AI as a artistic device, however that generative AI will help criminals scale up fraud by supplying the large music catalogs a scheme must repay.
The discover cites different current fallout too. Common Music Group sued distributor DistroKid in September over what it known as an “AI slop pipeline.” IFPI launched a Streaming Integrity Initiative the identical month, constructed round buyer verification, content material vetting and intelligence sharing.
What platforms are already doing
Streaming providers and distributors haven’t been idle. The discover describes platform-side detection that appears at listening time, geography and the ratio of saves and follows to streams.
Nonetheless, the Workplace notes that some scammers preserve their schemes alive by hopping from one distributor to the following as soon as they’re caught. One distributor has even floated the thought of a worldwide database of identified abusers.
There’s additionally a darker thread. The discover cites a Swedish investigative report alleging that felony networks used false streams to launder cash, and David M. Israelite, President and CEO of the Nationwide Music Publishers’ Affiliation, known as streaming fraud a “nationwide safety subject.”
What the Workplace desires to listen to
The inquiry poses 10 broad questions. Amongst them: how fraud is carried out and the way widespread every methodology is, what trade practices may be unintentionally enabling it, what the financial injury appears like, and the way fraud connects to AI-generated music and to dwell efficiency income.
The Workplace additionally desires to know what has really labored, together with trusted-distributor applications, royalty clawbacks, fines and content material removing. It’s asking about shared fraud databases, metadata standardization and whether or not to maneuver away from the pro-rata payout mannequin towards an artist-centric or user-centric one.
On the coverage aspect, commenters are invited to weigh in on laws equivalent to no-fault injunctive aid, a treatment that seems in a number of payments within the 119th Congress, together with adjustments to information-sharing guidelines and stronger felony enforcement on the state and native stage.
weigh in
Preliminary feedback are due by 11:59pm ET on November 23, 2026 and reply feedback are due December 21. Submissions undergo laws.gov and directions are posted on the Copyright Workplace’s web site.
There’s no assure any of this results in new legislation, but it surely’s a uncommon open invitation to form the file Congress can be working from.
