The Value of Art in the Age of Artificial Intelligence: Product or Experience?

Around two and a half thousand years ago, Plato described humans' tendency to mistake shadows for reality in his allegory of the cave. Today, the shadows on the cave wall have been replaced by screens. We live amid social media feeds, algorithms, artificial intelligence tools, and endlessly multiplying digital content. We are closer to information than ever before; yet perhaps for the first time, our relationship with reality has become this complicated.

And this complexity is no longer limited to news, images, or digital identities. It extends into fields we once thought belonged exclusively to human creativity. For many years, creative pursuits such as art, literature, and music were seen as the most authentic expressions of human experience. But in recent years, as artificial intelligence has begun producing impressive results in these areas too, old questions have returned in new forms.

Artificial intelligence can now write poetry, construct stories, generate images, and compose music. It is rapidly entering fields that, until just a few years ago, we thought belonged solely to human creativity. This is why the same question has been asked repeatedly in recent years: Can artificial intelligence make art? Yet the place these discussions around AI lead us may lie beyond that question, because the emergence of AI is prompting us to ask a far more fundamental question:

What is it that makes art art?

One of the names who has pursued this question is Margaret Atwood. Atwood has never seen creativity as merely the ability to produce words. For her, what makes a novel valuable is not the words on the page but the human experience behind those words. A writer is not simply the sum of what they have read. They are the carrier of the life they have lived, their fears, their losses, their desires, and their way of perceiving the world.

For Atwood, therefore, art is not simply the product that emerges. It is the invisible bond formed between the artist and the work. When we read a poem, we are not only reading the poem; we are also trying to understand how the person who wrote it sees the world. Inside a novel, we are not only looking for a plot; we are tracing the impressions of another human consciousness.

What Atwood has been pointing to in different ways for years is that people do not simply connect to stories — they also connect to the people telling those stories. When reading a novel, we do not merely follow events. We are curious about the eyes that saw that world, the mind that formed those sentences, and the life behind those lines. Perhaps this is why artificial intelligence can produce the form of a text, but cannot produce the relationship a human being has with the world.

Artificial intelligence, however, stands in a different place here. Large language models can process billions of texts, imitate styles, and reproduce forms — but they have no past, no childhood, no loss, no fear, no death. They have not passed through grief, have not lived inside a love, have not lain awake through a night in fear of death. It is precisely for this reason that artificial intelligence can imitate the form of art but cannot produce the human experience that is art's source.

At this point, the evaluations made by theater director Ege Maltepe on the play Shakespeare Shakspere mi? offer a striking perspective. Maltepe says, "A Hamlet learned from artificial intelligence is something like a 3D-printed copy of a real human being." This comparison may seem harsh at first glance, but the question beneath it is quite significant: Is truly knowing a work the same as having information about it?

Learning about Shakespeare from a summary is not the same experience as getting lost in the pages of Hamlet, because art most often creates transformation rather than transmitting information. What changes us is not only the work itself but the long and personal relationship we build with it.

Reading a book is not the same as knowing its summary. Experiencing a work of art is not the same as having information about it. Perhaps one of the greatest illusions of the AI age emerges here. We are beginning to substitute knowledge for experience. We can speak about a novel without reading it, interpret a painting without seeing it, take ownership of an idea without wrestling with it. If we recall Jean Baudrillard's theory of simulation, the problem is not merely the multiplication of copies. The problem is that copies begin to take the place of the real. We may be at a similar threshold in the domain of art.

In fact, the art world had begun encountering these questions long before artificial intelligence's rise in recent years. Artists have been exploring the role machines might play in the creative process for decades. In the 1970s, Harold Cohen's AARON system was one of the first examples of a computer producing drawings within its own set of rules. Later, Mario Klingemann produced portraits and constantly shifting visual identities with AI, while Sougwen Chung explored spaces where humans and machines could create together. Trevor Paglen and Hito Steyerl turned their attention not so much to what the machine produces as to how it sees and classifies the world. Although AI-assisted art has become more visible today through figures like Refik Anadol, it has in fact been drawing us toward the same question all along: Is it production or meaning that transforms something into a work of art?

Today, images produced with artificial intelligence can be shared millions of times. Hundreds of poems can be written in a few seconds. An endless number of stories can be generated. But in the middle of all this abundance of production, another question appears:

If everything can be produced, where is value created?

One of the most visible examples of this question appears today in the work of Refik Anadol. Anadol reprocesses millions of pieces of visual data, museum archives, and nature-related datasets through AI systems to produce large-scale visual experiences. His Unsupervised installation shown at MoMA in particular was interpreted as a museum collection reimagined by a machine. Perhaps it is exactly at this point that the discussion moves away from technology and returns to the human. The questions the art world asks are as striking as the wonder audiences experience before these works: Is it the algorithm itself that moves us, or the human intention that transforms that algorithm into an artistic practice?

Perhaps this is precisely why we are so astonished in the face of artificial intelligence today, because for the first time we are confronted with a system that produces outputs resembling human creativity but does not possess human experience.

Artificial intelligence can write poetry, construct stories, produce paintings, and even at times move us with the results it generates. But while doing all of this, it does not fall in love, does not grieve, does not feel longing, or face the fear of death. What it produces may resemble human experience, but it does not arise from that experience itself. Perhaps this is why artificial intelligence produces not experience, but the trace of experience.

It can construct the form of a love poem, but it does not fall in love. It can imitate the rhythm of a lament, but it does not lose. It can create a character, but it does not live. What emerges can sometimes resemble the experience itself to a surprising degree. Yet between resemblance and lived experience, there remains a distance that cannot yet be crossed.

This distinction may seem small, but it touches on one of the central questions in art, because when we look at art history we see that people have valued not only aesthetic forms but also the lived life behind those forms. Perhaps the very tension at the center of today's AI debates lies here. On one side are increasingly perfect imitations; on the other, the traces left by a life actually lived.

Artificial intelligence can write like Shakespeare, produce paintings reminiscent of Van Gogh, or approach the rhythm of Özdemir Asaf. But when we look at art history, we see that people have valued not only forms but also lives, because works of art are most often not merely aesthetic objects — they are also records of human experience. Perhaps the art debates of the future will take shape not around technical competence but around meaning, because the issue is no longer whether AI can produce, but why it produces.

Today, it is becoming less and less surprising that an artificial intelligence can write a poem or produce a painting. With every new model, technical limits are pushed a little further. But as technical competence increases, another question comes to the fore:

Is producing the same thing as creating?

If what makes a poem a poem is not merely the arrangement of words side by side, and if what makes a painting valuable is not merely the harmony of colors, then where should we look for art's true source?

Perhaps art's value lies not in the outcome that emerges but in the place from which that outcome arose. When a human writes, they are not merely producing text. They are trying to understand the world. Trying to understand themselves. Trying to preserve what they have lost, transform their fears, and build invisible connections with others. This is why art is most often not a product but a search for meaning.

Artificial intelligence, on the other hand, can produce things that resemble the outcome of this search. It can construct the form of a love poem, imitate the language of grief, reconstruct the structure of a novel. But while doing so, it does not live the search for meaning itself. Perhaps the most important difference between human production and AI production arises from the fact that one is born of experience, while the other is born of the traces of experience.

Refik Anadol's data landscapes, Sougwen Chung's human-machine co-productions, Mario Klingemann's uncanny portraits, and Hito Steyerl's critiques of digital imagery may seem like very different works. But all of them circle around the same question: Is what makes a work valuable only the outcome that emerges, or is it the meaning that outcome carries? The rise of AI-assisted art is perhaps important for this reason, because these works give us not the end of art but the opportunity to ask anew what art is.

At this point, an important distinction must be made between artists who make art with AI and users who merely ask AI to produce something art-like via a prompt. This distinction is important not to diminish AI or to devalue prompt-writing, but to see more clearly where artistic practice begins.

An artist working with AI most often does not simply want an image. They build a system, select a dataset, define an aesthetic problem, construct a conceptual framework, and situate the resulting output within a broader intellectual context. Refik Anadol's relationship with datasets, museum archives, and architectural spaces; Sougwen Chung's investigation of the shared movement of the human hand and robotic systems; Trevor Paglen's questioning of how machines see and classify the world — these cannot simply be read as 'images produced with AI.' At the center of these works is not the tool but the question of what the tool makes visible.

By contrast, simply telling an AI tool to 'paint in the style of Van Gogh,' 'write a melancholic love poem,' or 'create a futuristic cityscape' does not always amount to the same level of artistic practice. Such a prompt can certainly contain an idea, a wish, or a visual intention. But it most often does not carry the artist's sustained research, their relationship with materials, their process of trial and error, their conceptual responsibility, and how they situate the resulting work in the world. The difference here lies not only in the quality of the output but in the intellectual depth behind the production.

From the perspective of an art specialist, what makes an AI-generated image art is not simply its being impressive. What matters is what question it emerged from, in what context it was produced, what it transforms, and what kind of meaningful relationship it establishes with the viewer. Throughout art history, every new tool has generated a similar debate. When photography arrived, painting did not end — rather, painting was forced to rethink its own possibilities. Video, performance, installation, and digital media also expanded art's materials without reducing art to the tool alone. This is why the essential distinction in AI art is far more complex than the question 'did a human make it, or did a machine?' The more important question is: Where is the human in this process? Are they only the person who requested the result, or the person who built the system, defined the question, selected the materials, interpreted the outcome, and situated the resulting work in a meaningful context?

Prompt-writing can be part of the creative process. When used within a strong artistic practice, it can even function like an artist's sketchbook, camera, brush, or editing table. But prompt-writing alone is most often not enough to carry all the creative responsibility of a work, because art is not only producing a result — it is also constructing the reason for that result to exist.

Perhaps the fundamental difference that separates the artist from the prompt user in the age of AI emerges here. The artist does not simply receive an output from AI; they build a problem with it. They incorporate the tool into their own thinking, investigate its limits, see its errors and possibilities, and relate what emerges to their own world, to art history, and to the viewer's experience. That is, they use AI not merely as a production machine but as an intellectual material. This is why making art with AI is different from having AI produce something that looks like art. The first can open a space for aesthetic and intellectual inquiry. The second can most often remain an impressive, fast, and superficially successful production. The difference, again, comes back to where we have been throughout this piece: What transforms something into a work of art is not only production — it is meaning.

When we look at art history, we see that most great works have endured not because of their technical perfection but because they made human experience visible. This is why we still read Shakespeare, Dostoevsky, Virginia Woolf, or Özdemir Asaf today. They left us not only texts but also a way of living in the world as a human being. Perhaps it is precisely here that the role of the human becomes visible again in the age of AI. Artificial intelligence can be used as a powerful tool at many stages of the creative process. It can accelerate research, offer new alternatives, and expand the artist's field of work — but it is still a human who decides what is worth telling, which story to pursue, and which emotion carries meaning, because it is still a human who lives through pain, feels love, carries the weight of loss, and faces the idea of death.

We keep coming back to the same question: In the age of AI, the real issue is not whether AI can make art, but what makes art possible.

If art were only a matter of technical skill, this debate would have ended long ago with the advance of artificial intelligence. But art has never been only technical skill. It has always been an attempt to produce meaning, an expression of the human relationship with the world, and an effort to understand oneself as much as others. This is why artificial intelligence perhaps does not bring art to an end — on the contrary, it enables us to think again about what art is. With the rise of AI, what makes a poem unforgettable, what keeps a novel alive even decades later, or what distinguishes a painting from millions of other images becomes visible again. The answers to these questions lead us not to technology but to the human, because when we follow those questions honestly, we keep arriving at the same place: At the center of art is not technology, but the human.

Art is, in the end, a human being's effort to reach another human being. What makes a love poem unforgettable is not only the words; it is the longing, the loss, the waiting, and the vulnerability behind those words. What makes a lament lasting is not only its rhythm; it is the helplessness felt before death. For centuries, people have looked in works of art not for perfection but to find a part of themselves. Perhaps the most important distinction that the age of AI brings to light lies here. AI is learning to produce; it can write texts, generate images, and produce results resembling human creativity. But what transforms something into a work of art is not production — it is meaning. What makes a work unforgettable is not how it was made but why it exists, and that why still arises from human experience.