There’s a painting I think about all the time. It doesn’t hang in a museum. No placard, no donor’s name, no velvet rope. It lives on a screen, posted at 2:37 a.m. by an artist in São Paulo who’s never had a gallery show. A digital collage—layered, glitchy, half-abstract, half-figurative—and it stopped me cold. I found it because Instagram’s algorithm decided I should. Not a curator. Not a critic. A machine, trained on my lingering glances and double-taps, calculated that this image would hold my attention for an extra three seconds. It did. So what do we call that? Discovery? Patronage? Or something weirder: a new kind of gatekeeping dressed up in the cozy language of personalization.
The question isn’t whether social media platforms influence art. They do, obviously. The real question is how deep that influence goes—whether it’s just tweaking distribution, or quietly reshaping the thing we point to and say, “That’s art.” I’m not here to romanticize the old world. Galleries were never neutral; museums have always been political. But the algorithm is a different animal. It doesn’t pretend to have taste. It has metrics. And those metrics are starting to sculpt not just what we see, but what gets made in the first place.
The Feed as Curator
Let’s start with the obvious: the algorithm decides visibility. On Instagram, TikTok, even Pinterest, what surfaces is what performs. Performance means engagement—likes, shares, comments, saves, watch time. A quiet watercolor of an interior scene might be exquisite, but if it doesn’t trigger a quick reaction, it sinks. A neon-drenched, high-contrast digital portrait with a snappy caption? That travels. Artists notice. They adapt. Not always cynically; often just practically. You want your work to be seen. So you learn the language the platform rewards.
This isn’t exactly new. Artists have always responded to the demands of their patrons. The Medici family wanted certain kinds of religious imagery; the Dutch merchant class wanted still lifes and portraits. But those patrons were human beings with specific, often quirky desires. The algorithm is a composite of millions of desires, averaged into a statistical model. It doesn’t love anything. It doesn’t even prefer anything. It just amplifies what already generates a reaction, creating a feedback loop that narrows the aesthetic range over time.
I’ve watched painters shift from subtle palettes to saturated ones because muted tones don’t pop on a backlit phone screen. I’ve seen photographers abandon narrative series in favor of single, iconic images that work as thumbnails. Sculptors now think about how their work photographs from one specific angle—the angle that will become the post. The object itself becomes secondary to its documentation. That’s a profound shift. It’s not just about making art; it’s about making content that signals “art.”
The Rise of the Optimized Aesthetic
There’s a look that dominates certain corners of Instagram and TikTok right now. You know it: hyper-smooth gradients, surreal 3D renders that feel like candy-coated dreams, digital portraits with impossibly luminous skin, short video loops of glistening liquids or melting objects. It’s visually arresting, technically impressive, and engineered for the loop—designed to be watched three times in a row before the viewer even realizes they’re repeating. This is art that understands the assignment. The assignment is the algorithm.
I’m not knocking it. Some of this work is genuinely brilliant. Artists like Andreas Wannerstedt and Alexis Christodoulou have built careers on these hypnotic loops and pristine digital spaces. Their work is meticulous, conceptually rich, and deeply satisfying. But it also represents a narrowing of what “digital art” can look like when the primary distribution channel rewards a specific kind of visual pleasure. The algorithm loves smoothness. It loves immediate legibility. It loves a certain kind of prettiness that doesn’t ask too many questions. What it doesn’t love is ambiguity, slowness, or discomfort—qualities that have historically been pretty central to interesting art.
This creates a strange paradox. Social media has democratized access to an audience. Any artist with a phone can build a following. But the cost of that access is often a subtle conformity to the platform’s aesthetic preferences. The algorithm doesn’t censor; it simply doesn’t amplify. And in an attention economy, not being amplified is a kind of erasure. So we get a lot of art that looks like it was made to be scrolled past—and stopped, briefly, by the sheer force of its visual punch. Punchy art. Scroll-stopping art. Art that competes with memes, news headlines, and dance challenges. It’s a tough arena.
The Performance of Process
One of the savviest adaptations I’ve seen is the rise of process content. Artists don’t just post finished work anymore; they post themselves making it. Time-lapse videos of a mural coming together. Close-up shots of brushstrokes. The satisfying peel of masking tape. This content performs incredibly well because it satisfies the algorithm’s hunger for narrative and transformation—a beginning, middle, and end compressed into 30 seconds. It also humanizes the artist, building a parasocial bond that translates into sales and commissions. Smart. But it also changes the kind of work that gets made. If you know you’ll be filming the process, you might choose techniques that film well. Messy, slow, contemplative processes don’t make good Reels. Bold, gestural, high-contrast ones do.
I’ve talked to painters who now structure their studio days around content creation. They set up lighting and cameras before they set up their easels. They plan “reveal moments” in the painting process—points where the image suddenly coheres, perfect for that viral transition. The painting becomes a performance, and the performance becomes the product. The actual canvas, hanging on a collector’s wall, is almost a souvenir of the content. This isn’t a degradation of art. It’s a transformation. But we should be honest about what’s being transformed.
When the Crowd Becomes the Critic
Art criticism used to be a slow, elitist, often maddeningly opaque business. A few voices, mostly white, mostly male, mostly based in a handful of cities, decided what was serious and what was frivolous. That system deserved to be dismantled. Social media helped dismantle it. Now, anyone with an opinion can weigh in. The crowd is the critic. And the crowd’s judgment is instant, brutal, and numerical. You don’t get a thoughtful essay; you get a like count. You don’t get a layered debate; you get a comment section that veers from “fire emoji” to “this isn’t art” in three replies.
This has consequences for what artists are willing to risk. Controversial, challenging, or deeply personal work often gets a mixed reception. The algorithm reads mixed reception as low engagement and buries it. So artists self-censor, not because they’re afraid of criticism, but because they’re afraid of invisibility. The result is a public sphere of art that skews toward the broadly palatable, the easily digestible, the emotionally unambiguous. We’re losing the weird edges. The difficult corners. The work that makes you argue with yourself.
I think about the Dadaists, who would have been absolutely obliterated by Instagram’s algorithm. Their readymades, their nonsense poems, their anti-art provocations—none of it would have survived the engagement metrics. It would have been too strange, too off-putting, too slow to “get.” And yet Dada reshaped the entire 20th century. What are we missing now because it can’t get past the first round of algorithmic filtering? What kind of art never even gets made because the artist knows, intuitively, that it won’t travel?
The Meme as Medium
But let’s not be too precious. The algorithm has also birthed genuinely new forms. The meme, for instance, is arguably the most significant artistic medium of the 21st century. It’s collaborative, iterative, anonymous, and deeply responsive to the cultural moment. A single image macro can carry more political and emotional weight than a whole gallery of paintings. Memes are art that understands distribution as part of its DNA. They’re not just shared; they’re remixed, mutated, and weaponized. The algorithm loves them because they’re high-velocity. But they’re also, at their best, a kind of folk art—a vernacular visual language that belongs to everyone and no one.
Artists like Molly Soda and Brad Troemel have been playing in this space for years, treating the feed as both studio and exhibition venue. Their work is slippery. It looks like content, acts like content, but asks questions that content usually doesn’t. It’s art that wears a disguise. And that disguise is necessary because the algorithm doesn’t have a category for “critical interrogation of the platform itself.” It just sees an image. If the image performs, it spreads. The critique spreads with it, hidden in plain sight. That’s a kind of genius.
The New Gatekeepers
So who, or what, is the gatekeeper now? It’s not a person. It’s not even a single algorithm. It’s a system: the interplay of platform design, user behavior, and the economic incentives of attention capitalism. The algorithm is just the most visible part of that system. Beneath it are the engineers who tune the weights, the product managers who set the KPIs, the advertisers who fund the whole thing. They don’t think of themselves as art world players. But they are. Their decisions about what counts as “meaningful engagement” ripple through studios and galleries and art schools. A tweak to the recommendation model can shift the visual culture of millions of people.
This is power without accountability. The old gatekeepers—critics, curators, collectors—were at least visible. You could argue with them. You could understand their biases. The new system is opaque. Nobody outside a few tech companies knows exactly why certain images surface and others don’t. Artists are left to reverse-engineer the rules through trial and error, like gamblers trying to beat a slot machine. Some succeed. Many burn out. The psychological toll is real. I’ve heard artists describe the feed as a “dopamine casino” that rewards them unpredictably, keeping them hooked on the next post, the next hit of validation. That’s not a healthy creative environment.
And yet, there’s no going back. The gallery system was never going to be the primary way people encounter art in the 21st century. The screens are too ubiquitous, the attention too fragmented. The challenge isn’t to escape the algorithm—it’s to understand it well enough to subvert it, to build alternative spaces within and alongside it, to insist on values that can’t be reduced to engagement metrics. Slowness. Difficulty. Weirdness. Silence. These are not anti-algorithmic; they’re just not optimized. And maybe that’s the point.
FAQ
Does social media actually change what kind of art gets made, or just what gets seen?
Both, and they feed each other. Artists see what performs well and adjust their practice—sometimes consciously, sometimes not. Over time, the visible becomes the desirable, and the desirable becomes the norm. It’s a slow aesthetic drift, not a sudden revolution. But the cumulative effect is real: certain colors, compositions, and formats dominate because they’re platform-friendly.
Is algorithm-driven art inherently less valuable or serious?
Not at all. Value and seriousness are human judgments, not algorithmic ones. Some of the most exciting work today is being made in direct dialogue with the platforms that distribute it. The question isn’t whether the work is “serious” but whether it’s aware of the forces shaping it. Uncritical embrace of the algorithm is a problem; critical engagement with it is a practice.
Can artists resist the algorithm without disappearing?
Yes, but it requires strategy. Some artists build small, loyal audiences on purpose, prioritizing depth over reach. Others use social media as a portfolio and sell through direct channels. Still others create work that’s deliberately anti-algorithmic as a statement. The key is to treat the platform as a tool, not a master—and to have a clear sense of what you’re unwilling to compromise, even if it costs you views.
What role do memes play in all this?
Memes are the native art form of the algorithmic feed. They’re born to spread, mutate, and die quickly. But they’re also a powerful vehicle for cultural commentary, political satire, and collective emotion. The best meme-makers are artists, whether they call themselves that or not. They understand rhythm, repetition, and the visual language of the internet at a bone-deep level.
In the end, the algorithm is just a mirror—but a distorted one. It reflects not what we truly value, but what we’re willing to pause for. The pause is a thin slice of attention, easily manipulated, easily measured. Art has always asked for more than a pause. It asks for contemplation, for return, for the slow build of meaning over time. The algorithm can’t measure that. So it pretends it doesn’t matter. But it does. It’s the whole point.
We need to keep making the work that doesn’t fit. The work that’s too quiet for the feed, too strange for the share button, too personal for the public performance. Not because it’s pure—nothing is pure—but because it keeps the ecosystem diverse. Monocultures collapse. Aesthetic monocultures collapse too. The algorithm wants a garden of identical flowers, all blooming on schedule, all optimized for the camera. Our job, as artists and as viewers, is to keep planting weeds.