Industry leaders have committed to a “zero-tolerance” approach to streaming fraud amid heightened concerns about the scale of the problem. With the explosion of AI-generated music on DSPs, there’s an even greater threat from streaming manipulation.
A legal case in the US last month is seen as the first criminal prosecution of AI-based streaming fraud. It involved thousands of fake songs streamed billions of times, generating more than $8 million in royalties over several years.
“We are now properly acknowledging that it is a problem – we haven’t been vocal enough about it,” said Simon Robson, president, EMEA, Recorded Music, Warner Music Group. “We need to get control of it as soon as possible, because it’s going to grow very strongly. Bad actors have seen weak links in the streaming music chain and are exploiting them.”
While DSPs are taking action to address the volume of ‘spammy’ content, vast quantities of AI-created music have been uploaded across platforms. These tracks can then be streamed artificially by bots and, at scale, by using streaming farms to inflate the play count and benefit from DSP royalty payments – depriving real artists of earnings from the royalty pool.
“There’s a finite pot of money and anything which is taken away by bad actors, organised crime, impacts the art of very hard-working artists and songwriters, so it is a major issue,” said Robson.
The scale of the challenge across the streaming ecosystem remains uncertain.
“No one knows exactly, which is part of the problem,” Dennis Kooker, president, global digital business at Sony Music, told Music Week. “But could somewhere between 5% and 10% of the streams be fraudulent? Yeah, I think that’s possible. Unfortunately, the way that the ecosystem works, for those who really want to exploit it, they have a lot of incentive to create a lot of streams.”
Deezer said streaming fraud accounted for 8% of streams in 2025, based on fake streams it detected and demonetised.
“Streaming fraud needs to be addressed at all levels of the music industry and we continue to do our part and collaborate with our partners to stay ahead of the issue,” said Thibault Roucou, director, royalties and reporting at Deezer.
Spotify said that less than 1% of its streams are artificial.
“We continue to invest significantly in both automated detection and specialist review to prevent, identify and mitigate this activity,” said a spokesperson.
Spotify is trialling a profile protection feature to stop fake streams being attached to established artists’ pages. It follows the roll-out of stronger protections for real artists last year.
We’re underestimating how clever these bad actors are and how they’re finding loopholes – we need something so much more robust
Simon Robson
“Artificial streaming is an industry-wide challenge that pre-dates AI, though AI is increasing the speed and ease with which bad actors can attempt stream manipulation,” said the spokesperson. When illegitimate plays are detected, Spotify removes them from stream counts and royalties are not paid.
Deezer launched the first tagging system for AI-produced music, which it excludes from playlists. Some 75,000 fully AI-generated tracks are uploaded every day – 44% of the total. Deezer found that 85% of streams generated by these AI tracks are fraudulent, which it excludes from royalty payments.
Robson said there is the potential for “industrialisation” of streaming fraud, which would reduce the pool of royalties.
“I also worry about the consumer experience,” said Kooker. “Consumers are not happy with the amount of AI spam that they get hit with. It impacts costs inside the ecosystem. There are lots of reasons why this is not good for anyone who’s interested in a healthy, positive, thriving music business.”
The negative impact of generative AI coincides with a wave of licensing. Majors are signing AI deals while also taking legal action where they see a commercial threat.
“Validating business models that fail to respect artists’ work and creativity – and promote the exponential growth of AI slop on streaming platforms – is a grave disservice to artists, songwriters and all of us who work in music,” said UMG CEO & chairman Sir Lucian Grainge earlier this year.
While UMG has signed deals with tech firms such as music generator Udio, chief digital officer & EVP, Michael Nash, has spoken of their “walled garden” approach, which requires that users of licensed AI platforms cannot take music created with that technology outside of the service.
In November 2025, Universal Music Group, Universal Music Publishing Group, Sony Music Entertainment, Sony Music Publishing, Warner Music Group, and Warner Chappell Music all struck separate AI licensing agreements with LA-based music technology company Klay Vision. In addition to Klay, Warner have struck deals with Udio, Suno and Stability. The major will litigate if AI companies are being “exploitative”.
“But where we do feel there’s common ground and can start creating a licensed framework, that’s what we’ve done,” added Robson. “We are trying to lead the way to create an environment where there is a licence situation, so it could be beneficial for artists and songwriters.”
The industry will now be watching how the majors align on AI licensing.
“We always want the ecosystem to be innovating and evolving,” said Kooker. “We want new products and features coming to market, but they have to be things that are positive additions. Licensing something that adds to the dilutive problem that we’re talking about around streaming manipulation and AI slop is not a good addition to the market, so we’re not in favour of products that are going to have those types of results.”
By artificially generating plays for manipulated or fake content, bad actors are directly siphoning away essential revenues from creators, rights-holders and retailers
Dr Jo Twist
The rise of streaming fraud was addressed at the IFPI Global Music event.
“It is an increasing threat to the music industry,” BPI CEO Dr Jo Twist told Music Week. “By artificially generating plays for manipulated or fake content, bad actors are directly siphoning away essential revenues from creators, rights-holders and retailers, undermining the global music economy that supports human artistry.”
Twist added: “While labels are taking robust measures and are leading from the front, we need all parts of the streaming value chain, led by the DSPs and supported by our policymakers, to be proactive in safeguarding IP and copyright and to closely monitor, prevent and take effective steps against all forms of fraudulent activity.”
The BPI is working with labels and the IFPI, exploring a range of potential enforcement solutions.
Merlin, the indies’ digital licensing agency, hired Sarah McNabb as its first director of content integrity.
“The scale of attempted fraud is significant, but it must be viewed alongside the industry’s increasingly sophisticated response,” she said. “Merlin is determined to be at the forefront of this effort, working with our DSP partners to ensure a zero-tolerance approach to illegitimate activity.”
While acknowledging the sheer scale of fraudulent AI-based content, McNabb suggested it is “largely being intercepted before it impacts royalties payable on legitimate activity”. But offenders can evolve their strategies.
“Across the industry, we have moved from illegitimate activity being observed as obvious spikes of single tracks to ‘industrialised anonymity’,” said McNabb. “Now, the concern is more of a constant low-level hum, with hundreds or thousands of low-quality, often GenAI tracks each being streamed a relatively insignificant number of times to stay under the radar and blend into the background hum of the long tail.”
Spotify said it had removed 75m ‘spammy’ tracks from the platform, while Apple Music demonetised about two billion fraudulent streams in total.
“By negotiating with responsible partners like ElevenLabs and Udio, we can better differentiate between commercially cleared GenAI content and unlicensed slop, and build out guardrails and policies with our members and partners accordingly,” added McNabb.
Licensing something that adds to the dilutive problem that we’re talking about around streaming manipulation and AI slop is not a good addition to the market
Dennis Kooker
Robson said key measures include authenticating streaming users and tools to confirm content is legitimate.
“We need to stop the loopholes, and then we need to disincentivise people from actually trying to do this in the first place,” he told Music Week. “We’re underestimating how clever these bad actors are and how they’re finding loopholes – we need something so much more robust.”
Spotify works with labels, distributors and the Music Fights Fraud Alliance to combat cross-platform fraud. In flagrant cases, it applies charges to labels and distributors.
Using its protection systems, Sony Music has requested takedowns of 135,000 AI deepfakes of its artists’ music.
“It’s using technology to be able to find bad actors,” said Kooker. “We are actively putting resources against identifying these problems in the market, building controls into our own supply chain – it is a zero-tolerance position.”
Sony has put “massive investment” into tackling fraud in plain sight.
“The reality is that piracy has moved into the legitimate market,” said Kooker. “You can create a lucrative business if you’re able to skim money off.”
Amid a wave of global success for UK talent, Robson said it was time to act.
“It’s a shame that I’m talking about the negative elements when there are so many positives,” he concluded. “But all those artists are losing out, to greater or lesser degrees, because of this situation.”
PHOTO: (L-R) Dennis Kooker, Dr Jo Twist, Simon Robson. Credit: Louise Haywood-Schiefer (Dr Jo Twist, BPI)
FOLLOW 
