From ‘Fake Drake’ to Suno & Udio Label Settlements

An anonymous TikTok user using the pseudonym Ghostwriter goes viral for creating the song “Heart On My Sleeve,” which uses AI to deepfake Drake and The Weeknd vocals. Often referred to as the “Fake Drake” song, it is the moment when most music fans and industry folks realize that the age of AI music has arrived.

Inspired by the positive reception of Midjourney, an image generator, on Discord, Suno — which started as an AI-powered service to sort and organize music catalogs before pivoting to generative AI music — opts to launch its first model on the message service, capitalizing on its user base. 

Tennessee governor Bill Lee signs the Ensuring Likeness Voice and Image Security (ELVIS) Act into law, which is designed to protect the state’s artists from AI deepfakes. The ELVIS Act replaces the state’s previous right of publicity law, which only included explicit protections for one’s “name, photograph, or likeness,” expanding protections to include voice- and AI-specific concerns.

Founded by former researchers from Google’s DeepMind AI division, Udio launches publicly with financial backing from a16z, United Masters, will.i.am and more. At launch, the powerful AI music model can generate entire songs at the click of a button. 

Sexyy Red and Drake’s collaborative “U My Everything” samples “BBL Drizzy” (originally generated on Udio by comedian King Willonius, then remixed by Metro Boomin). This makes “U My Everything” the first known release to sample an AI-generated song, setting a precedent for how AI songs should be cleared in the future.  

After first being proposed in April 2021, the European Council formally adopts the EU AI Act, which places legal and transparency regulations on tech companies and AI developers operating in Europe, including those working in the creative sector and music business. This includes the core requirement that companies using generative AI or foundation AI models, like OpenAI’s ChatGPT or Anthropic’s Claude, provide detailed summaries of any copyrighted works, including music, that they have used to train their systems. The legislation also requires content generated by AI to be clearly labeled, and tech companies to ensure that their systems cannot be used to generate illegal and infringing content.

Universal Music Group, Warner Music Group and Sony Music team up on nearly identical $500 million copyright infringement lawsuits against Suno and Udio, alleging that the AI platforms unlawfully copied the labels’ sound recordings to train their AI models to generate music that could “saturate the market with machine-generated content that will directly compete with” human artists.

(These lawsuits are ongoing, although a few of the parties settled out of court in late 2025.)

A bipartisan group of U.S. senators introduces the NO FAKES Act to protect artists and others from AI deepfakes and other nonconsensual replicas of their voices, images and likenesses. For the first time, the legislation would create federal intellectual property protections for the so-called right of publicity, which restricts how someone’s name, image, likeness and voice can be used without consent. 

When NO FAKES fails to pass in 2024, it is reintroduced in 2025; once again, it does not pass. The bill has yet to be reintroduced for a third time. 

Michael Smith, a 52-year-old North Carolina musician, is indicted by federal prosecutors over allegations that, beginning in 2017, he used AI to help create “hundreds of thousands” of songs and then used the AI tracks to earn more than $10 million in fraudulent streaming royalty payments.

Manhattan federal prosecutors charge Smith with three counts of wire fraud, wire fraud conspiracy and money laundering conspiracy. According to the indictment, Smith was aided by the CEO of an unnamed AI music company (Billboard later reports a close connection between Smith and Boomy CEO Alex Mitchell) as well as other co-conspirators in the U.S. and around the world.

Paris-based company Deezer releases the first-ever report concerning how much AI-generated music is being uploaded to streaming services every day, noting that its team found 10% of daily uploads are fully created with AI. In the proceeding months, Deezer continues to update its figure, most recently saying that nearly 40% daily uploads are fully AI. In an interview with Billboard, Deezer’s director of research Manuel Moussallam explains the exponentially climbing numbers, saying: “We were very conservative in the numbers we reported [in the beginning]. We didn’t want any false positives.” Since then, he says the AI models have become more popular, and Deezer’s tool has improved. 

Hallwood Media, a management company, label and publisher, signs the first-known deal with an “AI music designer” — Hallwood’s preferred term for an “AI artist.” Called imoliver, the project was the brainchild of a human creator who uses Suno to create his work. He was scouted by A&Rs at Hallwood due to his popularity on Suno’s streaming service. 

Xania Monet, the AI artist project of Mississippi-born poet Telisha Jones, signs a multi-million dollar record deal with Hallwood Media. Best known for her gospel-inflected songs, including “How Was I Supposed to Know?” and “Let Go, Let God,” both of which have ranked on various Billboard gospel charts, Monet’s creator uses Suno to transform her poems into fully fledged songs. 

Spotify announces a number of strengthened policies around spam, impersonation and streaming fraud that target the challenges that come with AI-generated music. Because AI-generated music is made quickly and sometimes in the style of other artists, bad actors can use the tools to siphon money away from the royalty pool or to deepfake the identity of a human creative. While Spotify’s policies are not specific to AI-generated content — any form of impersonation, for example, would be penalized — it was the streamer’s way to rein in malicious users while allowing for the creative use of AI. 

Universal Music Group settles its part of the blockbuster copyright infringement lawsuit against Udio in exchange for a few important concessions. Most importantly, sometime in 2026, Udio will pivot its service from one that creates songs at the click of a button without licensing training material to one that allows fans to make remixes and mash-ups of fully licensed, pre-existing songs. 

Suno raises $250 million in a series C funding round — a sum nearly unheard of in the realm of music technology. The round is led by Menlo Ventures and also includes Hallwood Ventures, Lightspeed Ventures and Nvidia’s investment arm NVentures. 

Warner Music Group settles with Udio and Suno, effectively ending its part of both copyright infringement lawsuits filed with the other majors in summer 2024. Notably, Warner’s deal with Suno does not force the AI music company to make any stark changes, like Udio’s deal with UMG did. A press release states that sometime in 2026, Suno will relaunch with several changes, including  by training only on licensed material and placing a cap on user downloads. 

iHeartRadio announces its new program “Guaranteed Human,” pledging that the company doesn’t and won’t “use AI-generated personalities” or “play AI music that features synthetic vocalists pretending to be human,” among other promises. In a memo to staff, chief programming officer and president Tom Poleman writes, “‘Guaranteed Human’ is a core part of our brand. You’ll hear it in our imaging, and we want listeners to feel it every time they tune in.” iHeartRadio DJs must now add a line to their hourly legal IDs about being “Guaranteed Human.”

In a first for online listening platforms, Bandcamp places a ban on AI-generated songs, according to a company Reddit post. Specifically, Bandcamp is no longer permitting songs that use generative AI in all or a “substantial part” of their process on the platform. It also does not allow the use of any AI tools “to impersonate other artists or styles.”