Mikey Shulman, the CEO of music gen AI platform Suno, says that the music industry’s supposed resistance to AI innovation stems from a “fixed pie mentality” while admitting to using copyright protected music in his company’s AI training data, something that he describes as “stock standard” practice that “every AI company does”.
In a wide ranging interview on a leading venture capital industry podcast The Twenty Minute VC, Shulman tells host Harry Stebbings that making music is “not really enjoyable”, and that the time and practice required to master an instrument or production software is a barrier for most people. “I think the majority of people don’t enjoy the majority of the time they spend making music”, he says.
Throughout the interview, Shulman repeatedly emphasises the importance of human taste and judgment in music creation, arguing that “increasingly taste is the only thing that matters in art and skill is going to matter a lot less”, comparing this shift to historical changes in music culture: “Thirty years ago you wanted to be a rock star… Fifteen years ago you wanted to be a DJ… and now people want to be influencers”.
However, the Recording Industry Association Of America and major labels view Suno’s approach as potentially destructive rather than transformative. Shulman’s comments come as Suno faces litigation from the major labels, coordinated by the RIAA, which alleges unauthorised use of copyright protected recordings in the company’s AI training process.
The trade body is seeking damages of up to $150,000 for each copyrighted work used without permission, and has argued that unlicensed AI services like Suno “set back the promise of genuinely innovative AI for us all”, by claiming fair use under American copyright law to “copy an artist’s life’s work and exploit it for their own profit without consent or pay”.
Addressing the lawsuit indirectly, Shulman says “it just seems silly to throw a bunch of venture dollars at lawyers instead of sitting down and talking about how you could work together”. However, he acknowledges that the litigation poses a real threat: while insisting that “the company’s not dead” if the labels win their lawsuit, he admits that outcome would be “obviously not good for us”.
Shulman unsurprisingly takes issue with the record industry’s litigious approach. At the same time as admitting that his company’s AI model was trained on copyright protected music, Shulman insists that the industry should focus on collaboration rather than litigation. “The music business is one that has such an embedded fixed pie mentality”, he says. “There’s a fixed pie of money out there, and we are all just trying to divide it unfairly for ourselves”.
Despite facing potentially existential legal challenges, Shulman remains optimistic about AI’s role in music’s future. “The future is ours to build”, he says. “We can build a good future of music with AI and we can build a bad future of music with AI or we can sit back and let someone else do it”.
It’s clear in Shulman’s mind that Suno is the company to build that good future, noting that a core part of what drives the company is a cultural focus on music appreciation, driven in part by a decision to maintain a primarily in-person workforce at the company’s Cambridge, Massachusetts headquarters. “Every single person at Suno has, like, an incredible deep love and respect for music”, Shulman says.
Of course, the company’s staff is only one part of the equation – and in terms of costs, ranks far lower than GPU computing costs. That is “the biggest thing by far” in Suno’s operations, exceeding payroll “by a few times”, with the company’s significant investments in computing power spanning both production systems and what Shulman describes as “a big research cluster”.
As the music industry enters its next phase driven by “streaming 2.0” and the much-hyped superfan opportunity, Shulman outlines an ambitious vision for how generative AI might transform how people consume music, saying that his ideal future for music should be viewed through the lens of the video games industry. “Nobody half plays video games the same way people kind of put on music in the background”, he says. “I want to make music more like a video game”.
In fact, says Shulman, he thinks it’s critical that music should be more engaging and not just something that happens in the background, noting that he’s been knocking back requests to provide background music for other AI-generated content.
“The vast majority of gen AI video companies have asked us for an API”, he reveals. “The answer is always no… we are trying to make music more valuable for people and being the background music for your video is not making music more valuable”.
It is this more engaging approach to music, he argues, that could help the music industry match the games industry’s commercial success. “I’m sure I don’t need to tell you that the video game industry is so much bigger than the music industry”, he says, adding that “people have no problem parting with their hard-earned money to experience those things”.
Part of that experience, he suggests, is that rather than competing with artists, AI could enable new forms of fan engagement, pointing to a remix contest that Suno ran in conjunction with Timbaland as an example. “Getting to remix the music of your musical idol is like the ultimate form of engagement with them… so much cooler than, like, honestly even meeting them backstage after a concert”, adding that “the old model of stream share does not properly account for an interaction like that”.
While bigging up the opportunities for AI-powered music creation and fan engagement, Shulman claims that Suno prevents direct mimicry of existing artists. “You can’t make Ariana Grande songs”, he says. “If you put her name in, we’re gonna wave our finger at you and say that’s not what the future of music is”.
However, that statement is somewhat at odds with the RIAA’s characterisation of how Suno works, alleging in its lawsuit that Suno’s outputs demonstrate “strong resemblance to copyrighted recordings”.
The majority of artists Shulman talks to, he says, “use and love Suno”, adding that there are “only a couple” who are “not so into it”. This “biased sample” – by Shulmans’ own admission – is what gives him hope for the future.
“This makes me extremely optimistic about the future”, he says, claiming that artists may eventually embrace AI tools like Suno, with stars like Grande opting in the future to train personalised AI models on their own music, comparing the potential output to “fan fiction” for music. “I would love to get to a world where she can have models that make Ariana Grande songs”, he says, “an unlimited supply of future Ariana Grande songs in seconds”.
However, as anyone who has been following the music AI debate knows, this is far from straightforward, something that Shulman acknowledges, noting that artists “might not even be allowed to do that” depending on what their “contracts look like”.
Shulman is particularly scathing about current trends in pop music, arguing that songs have become “much more homogenous in melodies, harmonies, song structures”. He cites social media’s influence, noting how artists like Olivia Rodrigo are “changing the structure of songs to align to the algorithms of social channels”.
Though critics might note the irony of these observations coming from a company whose AI technology has been accused by the RIAA of potentially flooding the market with “machine-generated sound-a-likes and knockoffs”.
Looking ahead, Shulman outlines two scenarios he hopes to avoid. The first involves unregulated AI music creation by “a group in another country that doesn’t want to follow the laws” making unlimited unauthorised copies of artists’ styles.
The second is what he calls “hyper-personalisation” – where AI would generate endless streams of music perfectly tailored to individual listeners. “You open your phone… it knows everything about you… and it just streams you endless music that only you’re gonna like that is just hitting that nerve in your brain”, characterising this as “extremely antisocial” and counter to music’s communal nature.
When challenged about claims that AI would cheapen music by increasing supply, Shulman argues that while it might reduce “the average value of any given piece of content”, it will increase “the value of music to society or to any given person”.
Throughout the interview, Shulman is keen to emphasise that he believes that AI will ultimately increase music’s overall value, even if it potentially reduces the value of individual pieces of music. “I want a bigger, more vibrant music industry that has way more participants in it”, he says, “and not this more precious thing where very few people have access to actually doing it”.
That vision of democratisation of music creation enabled by AI appears to rely on a fundamental devaluation of musical skill and craft, however. “Increasingly taste is the only thing that matters in art and skill is going to matter a lot less”, he argues, suggesting that future success would come from choosing between AI-generated options rather than traditional musicianship.
For music creators – and the music industry more broadly – concerned about AI’s impact on their business, Shulman’s comments may do little to allay fears about companies like Suno. While claiming to want to make music “more valuable”, he simultaneously defends using copyrighted works without permission, dismisses the importance of musical skill, and suggests that traditional music-making is fundamentally unenjoyable for most people.
Shulman’s vision of Suno’s role in democratising music creation – “I’m not even sure you’ll know what model there is… you’re just gonna care that the music made you feel a certain way” – contrasts sharply with the industry’s concern about protecting artists’ rights and creative control.
His vision of AI democratising music creation appears to rest on contradictions: warning about unauthorised AI copying while admitting to it, criticising algorithmic influence on music while developing a platform that is ultimately powered by sophisticated algorithmic music generation, and claiming to support artists while using their work without permission in training data.
As one record industry executive noted in response to the RIAA lawsuit, “the winners of the streaming era worked cooperatively with artists and rightsholders to properly license music”.
Whether Suno’s approach of developing its technology first and seeking permission later will prove successful remains to be seen. This is, of course, an approach that has been tried and tested again and again over the years. With enormous computing costs and significant lawsuits against the company, the future for AI music – and the value that AI music may or may not generate for the wider industry – is far from assured.
Shulman’s comments reveal a striking disconnect between Suno’s stated aspirations and its actions. While claiming to want to “build a bigger and brighter future of music with AI”, the company chose to develop its technology using copyrighted works without permission, a decision that one of its investors acknowledged was a calculated risk, telling Rolling Stone that “If Suno had already done deals with labels at the point I invested… Honestly… I probably wouldn’t have invested in it”.
The Suno CEO’s characterisation of using copyrighted works as “stock standard” practice could prove particularly damaging in the context of the RIAA lawsuit, while his suggestion that “every AI company does it” could be interpreted as an admission of widespread copyright infringement across the AI industry rather than a defence of the practise in the context of Suno.
As the company awaits resolution of its legal challenges, Shulman maintains that Suno’s ultimate goal is to expand music creation beyond traditional barriers, bringing the ability to create music to more people than ever. “If you want to impact the way a billion people experience music you have to build something for a billion people”.
Whether Suno ultimately emerges as a pioneer of a new musical era or joins the list of tech companies that failed to survive music industry litigation will hinge almost entirely on how the courts view its fair use defence – and whether the company can convince the music community that its vision for music’s future is something the industry should support rather than oppose.