In 2018, the late avant-garde pop star and producer SOPHIE conducted her first-ever filmed interview. Her conversation partner, courtesy of Dazed magazine's "Age of AI" campaign, was Sophia, a speaking, articulating robot created by the Hong Kong company Hanson Robotics. The series was devoted to pursuing answers about AI's impending influence on artistic fields, and SOPHIE took the opportunity to ask Sophia about the technology's relationship to creativity and its impact on making music. "I believe we should be teaching AI to be creative, just as humans do for their children," the robot replied. "The best artificial intelligences in creative fields will probably be under partnerships with humans. … We don't truly understand the inner workings of the human mind, so I don't expect such a beautiful thing as creativity to be perfectly replicated. Just simulated." In 2025, Sophia's vision is starting to come into focus. Many operators have been working tirelessly in recent years to bring some kind of AI model to the pop sphere, and at the front of the pack this year is the rap producer Timbaland.
One of turn-of-the-millennium pop's big thinkers, Timbaland has co-launched a new AI-focused entertainment company called Stage Zero, which recently announced its first AI-generated "artist," TaTa. "Ultimately what Tim's here to do is to pioneer a new genre of music — A-pop, artificial pop," one of Stage Zero's co-founders, Rocky Mudaliar, told Rolling Stone. Their proposal conjures a kind of metaversal music — an omnigenre with no real roots, which can nonetheless interact with the real world. Timbaland in particular has been heralding the invention of a ground-up digital superstar. "I'm not just producing tracks anymore. I'm producing systems, stories, and stars from scratch," he said in a statement to Billboard. "[TaTa] is not an avatar. She is not a character. TaTa is a living, learning, autonomous music artist built with AI. … She's the first artist of a new generation. A-Pop is the next cultural evolution, and TaTa is its first icon."
The music for TaTa and fellow AI "artists" will be created using the music platform Suno, an AI music creation program positioned as a ChatGPT for song-makers. Like other forms of generative AI, Suno is designed to drum up new content based on what it has absorbed from existing media — inputting human-made works in bulk, then pulling from that massive data cache to fashion its output. Precedents for this tech have been stacking up lately: Recent digital tools can generate audio deepfakes that clone a singer's voice, and projects like Flow Machines have built AI tools for the automated composition of musical elements from a database, with dozens of presets for different genres. Suno is the next major step in that progression, transforming text prompts into fully realized songs, dubbed "generations," in under a minute. The TaTa character was derived from the platform's "Persona" feature, which allows users to capture the sound and voice profile from a particular generation and reuse it on future tracks. A song begins with Timbaland's demos being uploaded to the platform, which then spits out reworked versions in the likeness of other musical styles; human-written lyrics will be added after the fact. In a certain sense, there is an innovation to Suno, in that it effectively outsources production capabilities to assembly-line levels — but for many with a stake in music as a livelihood, it has been hard to see that change and its methods as anything other than erasure.
Timbaland's experiment comes at a particularly fraught moment for generative AI. At the end of last year, Spotify's annual Wrapped campaign was criticized for its inclusion of AI features. In January, Meta hastily removed tone-deaf AI-generated accounts from Facebook and Instagram. In April, the language-learning app Duolingo faced outrage when its company announced it would be transitioning into an AI-first business and eliminating contract work. SoundCloud fell under similar scrutiny last month when an eagle-eyed enthusiast noticed the platform's terms of service had been updated to allow the use of uploaded "content" as an input for AI models. And this month, backlash from editors prevented Wikipedia from adding AI-generated summaries for articles on the site. Following the trend, Timbaland's news was also met with disapproval. Longtime Jay-Z engineer Young Guru pleaded directly with the super-producer: "Your voice is powerful and way too important to do anything like this," he wrote in a reply on Instagram. "These are the times, right here, that history is defined.. Human expression can never be reduced to this!!!"
The tension surrounding AI's integration into artistic creation is driven by clashing interpretations of its application, where the utopian and dystopian visions can often feel overlaid on the same screen. In a perfect world, AI music platforms would allow for limitless exploration, expanding the existing capabilities of an imaginative person. That's the case that more intentional artists — such as the composer Holly Herndon, who branched out into new vocal techniques via a machine learning model she trained on her own voice — have made for and with their work. In theory, that should be the logical conclusion of the Timbaland vision, too. But so far, his execution lacks foresight. Days after the TaTa announcement, attention swarmed around a TikTok demo of his Suno process, in which he fed a beat produced by the artist K Fresh Music into the platform and turned around a flagrant facsimile of the original, all without permission from the creator. Everything promised by his new venture suggests a breakthrough in the most cynical sense: an advancement prioritizing ease of use, designed to make the old way extinct in its pursuit of maximum efficiency.
We already have examples of such efficiency disenfranchising artists and cheapening human expression. In 2022, the rapper-avatar FN Meka was signed to Capitol Records and dropped less than two weeks later, after onlookers raised questions about digital blackface. (As claims that the music and lyrics were generated by AI were challenged, additional questions surfaced about authenticity and creative control.) More recently, AI-assisted scammers have infiltrated Spotify, adding fake albums to artist pages and replica tracks to playlists, using the name recognition of existing stars to commit fraud. As Amazon moves to incorporate Suno into its virtual assistant Alexa, it may not be unreasonable to fear a more widespread replacement of human-made music with derivative, prompt-generated "covers" that are essentially one-to-one remakes, made by nonexistent artists leveraging deep-rooted cultural forms.
Still, for all the outspoken denunciation, individual use of AI has rapidly become routine. In 2023, Pew Research studies found that only 18% of American adults said they had used ChatGPT, and 52% were more concerned than excited about the use of AI in daily life. By the end of 2024, a study from Gallup-Telescope held that an estimated 99% of Americans used at least one AI-enabled product within the week of being surveyed, with nearly two-thirds of users not even realizing they had done so. A lot of those respondents claimed to dislike the technology — but its widespread adoption implies a subtle incursion, and the implications of this rampant use are devastating. This is where the dystopian part of the vision becomes hard to dismiss. College students are using AI to cheat. Minors are using it to create sexual images of their classmates. Google's AI-generated summaries can spit out factually wrong "hallucinations," which are then used to self-diagnose medical conditions. Reddit moderators have noted an uptick of users suffering delusions from emotional enmeshment with chatbots. Several heads of industry have toyed with creating an imaginary staff, embracing the self-mirroring feedback loop of their output as a plus. That's to say nothing of the encroachments upon books, film and visual art. With both older and younger users, the stigma is being eroded by convenience and accessibility, and Suno is positioned to ingratiate itself to a world that is normalizing generative art.
As a creative ideal, "artificial pop" feels profoundly disconnected from passion or purpose. Speaking about the platform's societal function last year, Suno co-founder Mikey Shulman put forth a head-spinning theory of what music is for, saying that Suno would correct a "lopsided" imbalance between the people who listen to music and the people who make it. "The way we think about this is we're trying to get a billion people much more engaged with music than they are now," he said. That take appears to minimize listening as a primary form of engagement — but Shulman has also been dismissive of the inherent rewards of creating, saying in January, "I think the majority of people don't enjoy the majority of the time they spend making music."
Suno seems to be working toward turning song construction into a calculus, stripping it of any practical labor. The latest version of the platform added a Song Editor tool that lets a user upload demos up to eight minutes long for remixing within the platform — changing lyrics, warping musical elements and transposing from one genre to another. A new track can also be created simply by humming a melody or adding a text prompt. According to the Suno blog, the Song Editor also now has three "creative sliders" that let users "choose how weird, structured, or reference-driven your generations get." The big idea is that you can crank out a lot of ideas very quickly with little effort — a formula for more, not necessarily for advancement, and a new existential threat to an already besieged musician class. In 2022, 80% of artists on Spotify had fewer than 50 monthly listeners; imagine how the math might change with people generating hundreds of songs in a few minutes.
Timbaland has admitted that TaTa grew out of his own struggles with obsolescence. "I thought it was over," he said in March. "Music is a young sport. I am the best, right? One of the best producers ever. I can make the drums, but something about it don't hit the same way in this generation." He has called Suno "Baby Timbo" — as in a mini-me creative engine — and hailed it as a divine lifeline from writer's block: "My Thriller, to me, is this tool. God presented this tool to me." It's a staggering thought coming from a beatmaker whose out-there style pushed the boundaries of Top 40 radio many times over. Timbaland cut his teeth imagining new sonic worlds for the rap futurist Missy Elliott and the late R&B sylph Aaliyah. His extensive catalog includes all-star team-ups with Justin Timberlake and Jay-Z that bent or helped shape their zeitgeist-defining legacies, and one-offs with international avant-pop figures Björk and Hikaru Utada. He once prided himself on that singularity, telling The Guardian in 2006, "I don't really like to look back. I've got a vault full of, like, a thousand reels of music that's never been touched that I've made over the years. I just like to do fresh stuff." There's something defeating in watching a restless auteur essentially turn on infinite lives for his unfinished drafts. Perhaps he sees Suno as merely the natural next step in a future-minded career, but his portfolio is also the case in point: Personal style is produced not just by filtering references through one's own perspective, but through trial and error. Timbaland is great because his unique taste was forged through observation, absorption and practice.
Personal battles aside, Suno's next hurdle is the music industry establishment. The Recording Industry Association of America is organizing lawsuits from the three major labels against the platform and its competitor Udio, claiming "mass infringement of copyrighted sound recordings copied and exploited without permission by two multi-million-dollar music generation services." Suno does not dispute this, instead asserting that the process should be legal — a subset of an ongoing argument waged by many AI firms that training the tech on copyrighted material is "fair use," a necessity for its evolution. (In a bit of irony, after shrugging off Suno's unauthorized scraping of licensed music, Timbaland asked Rolling Stone writer Brian Hiatt not to publish the prompts he was feeding to the platform to make songs, for fear they might not be copyrighted.) Labels may cave if granted a big enough piece of the pie, but for the moment, the artists they serve aren't hearing it. Musicians continue to petition against AI's rampant implementation without safeguards, with a 2024 open letter from the Artist Rights Alliance calling the pattern an "assault on human creativity."
At the time of this writing, TaTa's sound is still a mystery. No music has been released yet, just an uncanny publicity image, rendering the cyberstar as a blank-faced, pink-haired woman with a tiny face tattoo. But if Suno's purpose is to be an allusion generator, it may not really matter if the result is listenable — or even that Timbo is feeding his own creations into the digital shredder to make confetti. Broadly speaking, it is still designed to repurpose preexisting labor toward limiting future human input. In a recent interview, OpenAI Creative Director Chad Nelson made the point that there are very few humans who are truly originators, and that "a lot of what we create is derivative" — which is true. But humans learn and grow in a way the models can't. On its current trajectory, the surest promise that generative AI offers is not creating new originators, but massively scaling up the production of derivative art.
Ideation is the result of cognition, and often requires collaboration — people putting their heads together to puzzle out an answer — which is part of what makes generative music feel so sinister. Timbaland says he will continue to work with other artists, but his adoption of AI can feel uncomfortably like a means to bypass the accountability of human interaction. There is something off about a man puppeteering a femme-presenting virtual idol, commodifying the female voice without having to pay (monetarily or creatively) for that labor. Extrapolate that dilemma in every direction imaginable: Labels cashing in on AI songs by dead artists, or adding a prompt for "vocals in the style of X artist" to avoid having to work with them directly. Appropriation is a nonstop battle; the work of marginalized artists will only become easier to exploit and misrepresent in that world, and the outcome will always lack soul. When Rolling Stone's Hiatt sent a song he made with Suno in the style of Mississippi Delta blues to Living Colour guitarist Vernon Reid, the musician made a pointed assessment: "The long-running dystopian ideal of separating difficult, messy, undesirable, and despised humanity from its creative output is at hand." The blues is a music of struggle, the product of what Reid called "historical human drama." To approximate that art entirely by proxy, without confronting the drama, is to desecrate it.
Up to this point, AI music has been broadly met with disinterest from the public. When a faceless character going by Ghostwriter977 went viral for using AI to fake a song with Drake and The Weeknd in 2023, the move was initially received as a prognosis for music's future, but the novelty quickly wore off. Last year, Drake himself used AI vocal filters to perform as the late Tupac Shakur and Snoop Dogg during his beef with Kendrick Lamar, but was quickly served a cease and desist by the Shakur estate, forcing him to take the song down. Both instances involved using AI to pantomime an existing artist, whereas it seems the parties behind Stage Zero are betting on their incarnation feeling unprecedented and therefore less cribbed, its perceived newness blotting out the odious mimicry blatant in those other cases. Only time will tell if the ploy works, but there is really no artist-friendly end justifying these means, not even in the founders' own words. "The artists of tomorrow won't just be human, they'll be IP, code, and robotics that are fully autonomous," Mudaliar told Billboard, mirroring Timbaland's rhetorical stretching of "autonomy" to the limits of meaning. The dream of "A-pop" holds no apparent regard for authorship or design signatures, for identity or perspective, for any of the things that constitute artistic freedom. Its omnivorous approach, in search of an everything box for pop songcraft, seems to see imagination itself as a hindrance.
When Sophia had her summit with SOPHIE in 2018, she asked the artist, "What do you see as the purpose of music?" SOPHIE's answer was insightful: "Interpreting and reforming vibrations; sending and receiving vibrations, Sophia." I'm fascinated by that response, because it could be interpreted so dramatically differently by a machine and a human. You could make a case that interpreting and reforming vibrations is exactly what Suno is doing in the literal sense. But that, ironically, is also what separates us from these machine models, which, for all their generating, aren't doing any imagining. When swept up in music, as a listener or maker or both, a human mind knows that the connections being made are not computational, but spiritual — not just inputs and outputs, but intuition, too.
Copyright 2025 NPR
