The music streaming world has been stuck in a rut for years—Spotify, Apple Music, and YouTube Music dominate, each locked in their own walled gardens, while users juggle multiple subscriptions and playlists that don't talk to each other. But what if the next breakthrough app doesn't come from a tech giant's engineering team, but from an AI assistant coding up something entirely new? A recent column by David Pierce in The Verge explores this exact possibility, examining how "vibe-coded" AI-generated applications could reshape music discovery and cross-platform sharing in ways the current streaming titans haven't managed—or perhaps haven't bothered—to deliver.
The challenge isn't just technical; it's philosophical. Pierce argues that what's missing isn't better algorithms or higher bitrates, but genuine social connection around music—the ability to follow friends across platforms, share discoveries seamlessly, and break free from algorithmic echo chambers. This optimization-over-openness approach has created an opportunity for disruption—one that might come from an unexpected direction. According to Pierce's analysis, the streaming services we rely on today have optimized for retention and revenue, not for the kind of open, interoperable music culture that thrived in earlier internet eras. The question becomes: could an AI-built app, designed from scratch with modern needs in mind, actually deliver where incumbents have failed? And if so, what would that mean for how we discover, share, and experience music in the years ahead?
Why the streaming status quo is failing us
The fundamental problem with today's music streaming landscape isn't sound quality or catalog size—it's that platforms have become silos. Pierce points out that despite decades of digital music evolution, there's still no easy way to follow what your friends are listening to across different services, no universal "share" button that works whether you're on Spotify and they're on Apple Music, and no real social layer that connects music lovers the way early internet communities once did. Each platform wants to be your only platform, which means the social experience of music—once central to its appeal—has been sacrificed for corporate strategy.
The friction is real and immediate. Think about it: you discover an amazing track on Spotify and want to share it with a friend who's all-in on Apple Music. What happens? You send a link that either doesn't work properly, forces them to open a web player, or redirects them to sign up for a service they don't use. It's clunky, it's frustrating, and it breaks the spontaneous joy of music sharing that used to be as simple as handing someone a mixtape or burning them a CD (remember those?).
The column highlights that this isolation isn't accidental; it's by design. Streaming services keep users inside their ecosystems, where every play counts toward their metrics and every playlist keeps subscribers from churning—but this strategy clashes with how people actually experience music in 2025: fluidly, socially, and across multiple contexts. Pierce's research suggests that what users really want is interoperability—the ability to send a song to a friend without worrying whether they have the right subscription, or to see what's resonating in their broader social circles without platform barriers getting in the way.
The result is a paradox: we have more music available than ever before, yet discovery feels strangely isolated and algorithm-driven rather than human-powered. Your Discover Weekly playlist might be impressively accurate, but it's a fundamentally lonely experience compared to finding out what your actual friends are obsessing over. As Pierce notes, the streaming era has delivered convenience and catalog depth, but at the cost of the communal listening experiences that once defined music culture—from mixtapes to message board recommendations to simply knowing what your friends were spinning.
Could AI actually build the next iTunes?
Here's where things get interesting: what if the solution doesn't come from Spotify's next feature update or Apple's latest redesign, but from an entirely new app built by AI? Pierce explores the provocative idea that modern AI coding tools have reached a point where someone could theoretically prompt an assistant to create a music app that solves these exact problems—cross-platform compatibility, social discovery, and genuine interoperability—without the baggage of legacy systems or corporate incentives to maintain walled gardens.
It sounds almost too good to be true, right? You can rapidly prototype interfaces and integrations using modern AI coding assistants (examples: Claude Code, GitHub Copilot, etc.), though production-grade, secure, licensed services still require engineering and operational work. But we're closer to that reality than you might think. The barriers to building functional software have collapsed dramatically in the past couple of years, and what once required teams of engineers and months of development can now be prototyped in days or even hours.
The term "vibe-coded" captures something essential about this approach: rather than engineering an app through traditional development cycles, committees, and feature roadmaps, you'd be describing the experience you want—the vibe, the feeling, the social dynamics—and letting AI figure out the technical implementation. A developer could, in theory, tell an AI: "Build me an app where I can see what my friends are listening to across all streaming platforms, where sharing a song automatically opens in whatever app the recipient uses, and where I can follow tastemakers without caring which service they're on." The AI would parse those requirements, generate the necessary code, handle the UI, and potentially even suggest architectural patterns the developer hadn't considered.
According to Pierce's analysis, this isn't pure speculation; AI coding assistants are already capable of generating functional applications from natural language descriptions, and the gap between "proof of concept" and "viable product" is narrowing rapidly.
But there's a catch, and it's a big one: building an app is one thing; building a sustainable business that can license music, negotiate with labels, and compete with billion-dollar platforms is something else entirely. Pierce acknowledges that even the most elegantly designed, AI-generated music app would face brutal economics and licensing complexities that have kept new entrants out of the streaming market for years. You can code up the slickest interface imaginable, but without securing streaming licenses from the major labels—who aren't exactly eager to enable new players that might disrupt the comfortable oligopoly they've established with Spotify, Apple, and Amazon—you're dead in the water. The technology might be democratized, but the music industry gatekeepers decidedly are not.
Still, the possibility is tantalizing. The column suggests that AI-assisted development could enable rapid experimentation with new social features, interface paradigms, and cross-platform integrations that established players are too slow or too risk-averse to attempt. If someone—or some team—could crack the licensing puzzle, an AI-built app designed around genuine user needs rather than corporate metrics could genuinely disrupt the streaming status quo.
What a truly social music platform would look like
So what would this hypothetical "next iTunes" actually do differently? Pierce's vision centers on breaking down platform barriers and restoring the social dimension to music discovery. Imagine being able to follow friends' listening habits regardless of which service they use, seeing a unified feed of what people in your network are discovering, and sharing tracks that automatically open in whatever app the recipient prefers—Spotify, Apple Music, Tidal, or even YouTube.
Let's break it down. The core idea is that your music identity shouldn't be fragmented across different platforms. Your taste, your listening history, your social connections around music—these should be portable and platform-agnostic. You'd have one identity for music discovery that works everywhere, rather than separate profiles on each streaming service that don't communicate with each other.
The column emphasizes that this isn't about replacing streaming services, but about creating a layer on top of them—a social discovery engine that treats platforms as interchangeable backends rather than competing destinations. Think of it as a music-focused social network that's agnostic about where the actual audio comes from, prioritizing connection and discovery over lock-in and exclusivity.
The analogy that comes to mind is how email works: you don't need to care whether your friend uses Gmail, Outlook, or some obscure email provider because the protocols are open and interoperable. The same should be true for music sharing. The platform you happen to use for playback shouldn't dictate who you can share with or follow.
According to Pierce's framework, such a platform would also need to rethink recommendations entirely. Instead of purely algorithmic suggestions based on listening history, it would surface music through trusted human filters—friends, tastemakers, communities—while still leveraging AI to help you navigate and organize those social signals. The goal would be to combine the scale and personalization of modern algorithms with the authenticity and serendipity of human curation, something current streaming services have struggled to balance.
Here's what's interesting about that approach: algorithms are great at pattern matching and showing you more of what you already like, but they're terrible at the kind of surprising, context-dependent recommendations that come from actual humans who know you. Your friend who knows you went through a messy breakup might share a specific album at exactly the right moment in a way no algorithm ever could. That human intelligence layer has been systematically stripped out of modern music discovery, replaced by machine learning models optimized for engagement metrics rather than meaningful connection.
The technical challenges are real—API access, authentication across platforms, handling different catalog availability by region—but none are insurmountable. Pierce suggests that the bigger obstacle is cultural and economic: convincing streaming platforms to allow this kind of interoperability, and building a business model that works when you're not controlling the entire user experience from playback to payment. And that's the rub. Spotify and Apple Music have no incentive to enable this kind of openness. Why would they help build the infrastructure that makes it easier for users to switch between platforms or reduce their lock-in? The current model—where your playlists, social connections, and listening history are trapped inside one ecosystem—is working perfectly well for them, even if it's frustrating for users.
Where does music streaming go from here?
Whether or not an AI-coded app becomes the "next iTunes," the conversation Pierce has started points to a genuine gap in today's streaming ecosystem—and to growing user frustration with platform lock-in and algorithmic isolation. The column makes clear that the technology exists to build something better; what's missing is the will, the business model, and perhaps the right catalyst to make it happen.
The key takeaway is this: we're living in a moment where the tools for creating sophisticated software have been democratized in ways that would have seemed impossible just a few years ago. Someone sitting in their bedroom with access to AI coding assistants has capabilities that once required entire engineering departments. That's genuinely transformative, even if the path from prototype to viable business remains treacherous. Pierce's analysis suggests that we're at an inflection point where AI development tools could genuinely democratize app creation, enabling new kinds of music experiences that weren't feasible—or economically viable—just a few years ago. But it also highlights the stubborn realities of the music industry: licensing complexity, platform incentives, and the enormous moat that established streaming services have built around their user bases.
What's fascinating is how this mirrors broader tensions in tech right now. On one hand, we have increasingly powerful tools that enable individual creators and small teams to build sophisticated products. On the other hand, we have entrenched platforms with massive network effects and business models predicated on maintaining closed ecosystems—creating a dynamic tension between democratized creation tools and concentrated platform power. This same pattern is playing out across social media, app stores, and digital marketplaces, where the question of whether innovation or incumbency wins out will shape the next decade of technology.
The bottom line? As Pierce concludes, the next breakthrough in music streaming probably won't come from incremental feature updates or slightly better playlists. It'll come from someone—or some team empowered by AI—willing to rethink the entire experience from the ground up, prioritizing social connection and cross-platform openness over the walled-garden strategies that have defined the streaming era so far. Whether that happens through a vibe-coded startup, a platform finally embracing interoperability, or some hybrid approach we haven't imagined yet remains to be seen. But one thing is certain: the current state of music streaming isn't the final answer, and the tools to build something better are already here.
The question now is whether someone with the right combination of technical capability, industry connections, and sheer determination will actually pull it off—and whether the music industry's gatekeepers will let them.
Comments
Be the first, drop a comment!