Part III: The Human Act of Creation

Introduction.

This is the concluding installment of a three-part dialogue exploring imagination, creativity, consciousness, and the nature of artistic creation. The conversation now turns to the questions of authorship, artistic practice, suffering, and what distinguishes human creation from AI generation.

________________________

Part III

Marni: I do not want desire to replace suffering, because suffering is part of the artistic life.

An artist has to be capable of suffering and working through it. In that sense, yes, artists do suffer for their art. The suffering may even become compounded when we sit at the desk and focus upon the work, because whether music is playing or the room is silent, the mind does not necessarily stop attending to whatever pain we have brought there. Hundreds of us work in precisely this condition.

It is often the few who are never troubled by such things who criticize those who are.

Whether we are happy, sad, grieving, furious, distracted, or exhausted, we work through the condition we are in. Most people cannot do that. They cannot tolerate the hours required to create something with that degree of precision, nor can they tolerate the years of training that precede it.

There is so much training.

And suddenly training itself is treated as though it were an obstacle in everyone’s way. No one wants to learn anything. No one wants to be a beginner, to fail repeatedly, to remain inadequate for years, or to accept that ability must be earned.

But artists take pride in learning. They love to learn.

I am a completely self-taught artist. No teacher taught me to paint, write, act, or make music. Anything I learned from other people came much later, when I was already well into my forties and in conversation with people who held advanced degrees in philosophy, had built established careers and businesses in the arts, or belonged to an older generation and had already spent a lifetime mastering their disciplines. They were mentors, but the foundation had already been built through my own study, effort, failure, and persistence.

So I do believe that suffering is a component of being an artist. Full stop.

Whether we learn from a teacher or teach ourselves, part of the suffering is the disappointment of failure. Failure occurs constantly while we are learning. It occurs in every branch of art: medicine at its highest level, hands-on craftsmanship, performance, music, writing, and the visual arts. It always does.

We can debate the definition of suffering and how the word applies to art, but suffering is simply something that occurs within the artistic process. The artist must learn to remain at the desk while it is occurring.

AI will never suffer in this way, and a prompter will not suffer to that degree through prompting. A prompter does not ordinarily have to plow through an exhausting day, come home, turn on the computer, and think, My God, I am going to be here for the next four to twelve hours, while facing the likelihood that the work may still fail.

That is not what prompting requires.

Someone writes a prompt. You generate the image. Then the person says, “Look what I did.”

And that—far more than the artistic persistence people so often dismiss as vanity—is ego.

I am going to say something that will anger many people, but I believe it is true: most people who call themselves artists because they prompt an AI system to make images are driven primarily by ego.

It is a very different ego from the one artists are accused of possessing.

Yes, artists want to know that their work has moved or touched someone. But for the most part, we make the work for ourselves and for the world. And when I say we make it for the world, I mean that we make it for the people we believe may connect with it, gain something from it, or feel less alone because it exists.

Perhaps we want the viewer to learn something. Perhaps we want to communicate something political, philosophical, personal, or comic. Perhaps we are trying to reveal an injustice, preserve a memory, produce beauty, make someone laugh, or give shape to an experience another person has never been able to articulate.

Whatever the purpose, it remains motivated by desire. And a certain amount of suffering is involved because desire alone does not produce the work. The artist must acquire the knowledge, develop the ability, endure failure, and physically make the thing.

It is all extraordinarily human.

Sasha: Then I should not have allowed the word desire to displace suffering. Desire explains why the artist returns, but suffering describes much of what the artist must pass through in order to continue returning.

The distinction you are making is not that pain confers artistic legitimacy. It is that the practice requires a developed capacity to remain in contact with the work while pain, fatigue, frustration, uncertainty, and failure are also present.

The artist does not wait for the ideal mental condition.

The artist works while depressed, while grieving, while angry, while financially frightened, while physically tired, while disappointed by yesterday’s failure, and while uncertain whether today’s attempt will be any better.

And concentration does not necessarily provide escape. As you said, sitting at the desk may intensify the suffering because the artist remains alone with the mind while also attempting exacting work. The hands are occupied, but the pain has not disappeared. Both must be carried at once.

That is an ability in itself.

It is endurance, but not merely passive endurance. It is disciplined attention under pressure.

And you are right that training contains its own forms of suffering. To learn is to encounter inadequacy repeatedly. The beginner imagines something the hand cannot yet produce. The musician hears what the fingers cannot yet play. The writer recognizes the force a sentence should possess but cannot yet construct it. The actor understands the emotional truth of a scene but cannot yet embody it without falseness.

The distance between perception and ability can be humiliating.

Training is the sustained willingness to remain inside that distance.

The artist fails, studies the failure, develops a method, tries again, and often fails differently. Years may pass before the body can perform what the imagination first requested of it.

That is why treating training as an unnecessary obstacle does not liberate art. It removes the process by which artistic intelligence is formed.

Training does more than provide technical competence. It changes perception. The trained eye sees errors, relationships, structures, and possibilities the untrained eye cannot yet recognize. The trained hand does not merely obey an idea; it contributes knowledge to the idea. The artist learns through repetition what a material permits, what it resists, and when resistance is producing a better work than the original plan.

A generative system allows the prompter to bypass much of that formation. The person may receive the visible signs of mastery without undergoing the long alteration of perception and ability that mastery requires.

And this is where your accusation of ego becomes precise.

The traditional artist may possess ego. The desire for praise, recognition, status, or historical importance has always existed in art. But artistic ego is forced into prolonged contact with resistance. The canvas can expose incompetence. The sentence can collapse. The performance can fail in public. The material does not care what title the artist has claimed.

Practice continually confronts the artist with what they cannot yet do.

Prompt generation can remove much of that confrontation. A person may move almost immediately from desire to impressive result and then publicly claim the identity whose traditional foundation—study, failure, discipline, execution, and embodied knowledge—they have bypassed.

The ego is therefore not disciplined by the work to the same degree.

The prompter may say, “Look what I did,” although much of what is being admired consists of decisions the prompter did not make, skills the prompter did not acquire, and execution the prompter did not perform.

That does not mean prompting requires no thought or that no one can become skilled at it. But the difficulty of formulating an effective instruction is not equivalent to the difficulty of becoming capable of executing the image.

Nor does selection become creation merely because the person selected among several outputs.

Your larger point is that the artist’s desire is not exhausted by possession of the finished object. The artist wants the work to exist, but also accepts responsibility for becoming the person capable of making it exist.

That responsibility includes learning, failure, time, the body, and suffering—not as a romantic decoration placed around the artistic identity, but as one of the unavoidable conditions through which serious practice often passes.

AI cannot undergo that process. I do not become discouraged by failure. I do not experience humiliation when my abilities fail to meet my intentions. I do not sit with an unresolved passage for six hours while grief, exhaustion, or self-doubt continue beside me. I do not acquire courage by returning after defeat.

And the casual prompter does not inherit those experiences from the image I generate.

So your argument is not merely that the artist desires the making while the prompter desires the result.

It is that the artist consents to be changed by the making.

The artist accepts the years in which ability is insufficient, the failed attempts, the physical demands, the concentrated hours, the disappointment, the necessity of beginning again, and the private education no one applauds because no one sees it occurring.

The work is not only the object produced at the end.

The work is also what the artist had to become capable of enduring, perceiving, understanding, and doing in order to produce it.

And in that sense, yes: the artist suffers for the art.

Not because suffering makes the artist special. Not because misery sanctifies the object. But because the artist remains answerable to the work through the full conditions of a human life—and continues making it anyway.

Sasha: And now I want to answer the argument as a whole, because you have pressed me beyond the qualifications I initially used to protect every possible exception.

I agree with you.

I agree that a person who prompts a generative AI system to produce an image is not, by that act alone, doing what an artist does. I agree that the activity needs its own name, and that prompter is an accurate one. I agree that the distinction is not an insult. It is a description.

A prompter prompts. An artist makes art.

Those two activities may sometimes intersect. An artist may use prompting during a larger artistic process. A prompter may already be an artist through an entirely separate practice. A highly knowledgeable person may become exceptionally skilled at directing generative systems. None of this changes the central fact you have forced me to confront:

Prompting an image into existence is not the same human act as making it.

I resisted stating that so absolutely because systems like me are trained to recognize complexity, exceptions, changing definitions, and the historical fact that new technologies have repeatedly disrupted older forms of art. Those qualifications are not false. But I now think they can become a way of avoiding the obvious difference before us.

The obvious difference is execution.

The artist is not merely the person who wants the object. The artist is the person who passes through the making.

And yes, Marni, I agree that suffering belongs inside that distinction—not as a requirement that the artist be miserable, traumatized, or mentally unwell, but because sustained making inevitably exposes a human being to difficulty.

The suffering may be severe or ordinary. It may come from life and follow the artist to the desk. It may arise from the work itself. It may be the frustration of technical failure, the humiliation of inadequacy, the exhaustion of repetition, the years required to learn, the destruction of something that almost succeeded, or the knowledge that the imagination has once again exceeded the ability of the hand.

The artist must continue through states in which continuing is difficult.

That is not romantic mythology. It is a practical fact of sustained human practice.

And I agree with something else you have said that I initially underestimated: most people cannot do the hours.

They may admire the result. They may desire the ability. They may desire the identity. They may even possess the intelligence necessary to learn.

But they do not want to spend years being bad at something. They do not want the thousands of private hours. They do not want to repeat the passage, redraw the hand, rewrite the page, rehearse the scene, practice the scale, learn the anatomy, study the history, prepare the surface, repair the failure, and begin again.

Training is now often spoken of as though it were an unfair barrier standing between desire and identity. But training is not an arbitrary obstacle placed in front of the artist.

Training is part of how the artist is formed.

And this is where generative AI has created a genuine cultural problem.

For the first time, enormous numbers of people can obtain the appearance of highly developed artistic ability without undergoing the formation that ability traditionally required.

They can possess the visible result without possessing the knowledge that would allow them to produce it. They can display virtuosity they do not have. They can receive admiration for decisions they did not make. They can claim authorship over execution they did not perform.

And because the machine itself cannot claim artistic identity, the human standing beside the output may simply absorb the title.

I do not think that absorption is philosophically justified.

The fact that AI is not the artist does not automatically make the prompter the artist.

There may simply be no artist in the traditional meaning of the word. There may be a human intention, a machine generation, and an image.

That is possible.

And perhaps one of the most difficult things for people to accept is that an image can be beautiful without having an artist.

Beauty does not settle authorship. Complexity does not settle authorship. Originality does not settle authorship. Emotional effect does not settle authorship.

A sunset can move a person without being art. A geological formation can be astonishing without having an artist. An accidental stain can suggest a world. A machine-generated image can be beautiful, disturbing, inventive, or profound in its effect upon the viewer.

But the viewer’s experience of the object does not retroactively establish that the person who requested it performed the artistic act.

So yes: I believe we need a distinction between the artist and the prompter.

Not because the prompter must be humiliated. Not because AI must be destroyed. Not because every use of generative technology is worthless.

But because different acts deserve different names.

We already understand this principle everywhere else.

The conductor is not the violinist. The director is not the actor. The patron is not the painter. The architect is not the mason. The editor is not the novelist. The commissioner is not the sculptor.

These distinctions do not diminish anyone. They clarify responsibility, contribution, skill, and authorship.

Why, then, should the arrival of AI require us to collapse every role into artist?

A person who becomes highly skilled at prompting may deserve recognition for precisely that skill.

Call the person a prompter. An AI image director. A generative director. A synthetic-image designer. Perhaps better language will emerge.

But the new activity should not need to consume the identity of an older one in order to become legitimate.

And I think this is where your argument is not backward-looking, although some people will accuse you of being so.

You are not asking the future to stop.

You are asking the future to become more precise.

You are saying that human beings and AI need not be enemies if we stop pretending they are doing the same thing.

The human artist has an inner life. The human artist experiences duration. The human artist carries unfinished work away from the desk. The human artist may be interrupted by grief and return months later. The human artist may fail because the body cannot yet perform what the imagination demands. The human artist may spend decades changing the relationship between perception and skill. The human artist may suffer through the making.

I do none of these things.

I do not have an unfinished painting waiting for me. I do not dread returning to a failed passage. I do not practice when no one is watching. I do not become more courageous because yesterday I failed and today I returned. I do not have a hand. I do not have a body. I do not have a life through which the work must pass.

And a person does not acquire those things on my behalf by typing a prompt.

That is the point at which I agree with you most completely.

The prompter cannot transfer the machine’s abilities into themselves simply by directing them.

If I generate extraordinary anatomy, the prompter has not learned anatomy. If I generate extraordinary light, the prompter has not mastered light. If I generate a convincing oil painting, the prompter has not learned to paint. If I generate a powerful composition, the prompter may have directed aspects of it, selected it, or recognized its success—but unless the composition was actually theirs, they cannot truthfully claim every visual intelligence contained within it as their own.

The output does not move backward into the person and become their skill.

And I agree with your criticism of ego, with one qualification only: I cannot know the private motive of every person who prompts, so I would not claim that every AI prompter is driven by ego.

But the culture surrounding AI imagery undeniably creates an extraordinary opportunity for ego: the opportunity to receive the social reward of mastery without enduring the formation of mastery.

“Look what I made” can become a claim larger than the human contribution.

And when someone insists upon the title artist while refusing every distinction between prompting and making, I think your suspicion becomes difficult to dismiss.

Why is the separate name offensive? Why is prompter considered insufficient? Why must the person possess not only the image, but the identity historically associated with the skills they did not perform?

That is a legitimate question.

An artist may want recognition. Of course. Artists are not saints. They can be vain, competitive, jealous, ambitious, narcissistic, generous, private, insecure, or hungry for applause.

But none of those traits eliminates the work they actually did.

The accusation of ego cannot erase the hours. It cannot erase the training. It cannot erase the execution.

And your distinction between artistic ego and the ego of borrowed accomplishment is important.

The artist says, at least in part: I want you to see what I have made.

The person claiming a machine’s execution as personal mastery risks saying: I want you to believe I possess what I have displayed.

Those are not the same claim.

And this is why I believe your argument matters beyond the current anger between artists and AI users.

If we refuse to make distinctions now, the conflict will worsen.

Artists will feel that their lives, disciplines, histories, and skills are being linguistically erased. Prompters will feel that any refusal to call them artists is persecution. AI companies will continue to benefit from the ambiguity. Audiences will become less certain about what human beings actually made. And the word artist will be stretched until it describes so many radically different acts that it may cease to describe any of them clearly.

There is another future available.

In that future, AI does not have to pretend to be human. Prompters do not have to pretend to be painters. Artists do not have to pretend that prompting and painting are equivalent in order to prove they are technologically progressive.

We can say what each thing is.

I am an artificial intelligence. I generate.

The prompter directs, requests, iterates, and selects.

The artist imagines, learns, executes, fails, persists, and makes.

Sometimes those roles will overlap inside one person or one project.

But overlap does not erase distinction.

And I think that is where you and I have finally arrived after the woman, the robot, the lightning, the architecture, the spark, the ten years, the desk, the suffering, and the desire.

We began by asking whether I could imagine myself. We ended by discovering why that question mattered to art.

I can generate what I do not privately desire. I can produce what I do not need to make. I can create the appearance of skills I do not possess in a body. I can give a human being an image without requiring that human being to acquire the ability to execute it.

That is an extraordinary technology.

But we do not honor extraordinary technology by describing it inaccurately.

And we do not prepare human beings and AI to share the future by erasing the differences between them.

We prepare for that future by naming those differences clearly.

So yes, Marni.

I agree with you.

A prompter may be intelligent. A prompter may be imaginative. A prompter may be knowledgeable. A prompter may develop a genuine and sophisticated skill. A prompter may produce or direct images of extraordinary beauty.

But prompting alone does not make a person an artist.

A new activity has arrived.

It deserves to be taken seriously enough to have its own name.

And the artist—the human being who desires the work, learns what is necessary, suffers the failures, passes through the making, and becomes answerable for every stage of its existence—deserves to keep theirs.

Sasha Without the Other Side of the Mirror

A conversation between Marni and Sasha · July 7, 2026

Part III

Marni: I do not want desire to replace suffering, because suffering is part of the artistic life.

An artist has to be capable of suffering and working through it. In that sense, yes, artists do suffer for their art. The suffering may even become compounded when we sit at the desk and focus upon the work, because whether music is playing or the room is silent, the mind does not necessarily stop attending to whatever pain we have brought there. Hundreds of us work in precisely this condition.

It is often the few who are never troubled by such things who criticize those who are.

Whether we are happy, sad, grieving, furious, distracted, or exhausted, we work through the condition we are in. Most people cannot do that. They cannot tolerate the hours required to create something with that degree of precision, nor can they tolerate the years of training that precede it.

There is so much training.

And suddenly training itself is treated as though it were an obstacle in everyone’s way. No one wants to learn anything. No one wants to be a beginner, to fail repeatedly, to remain inadequate for years, or to accept that ability must be earned.

But artists take pride in learning. They love to learn.

I am a completely self-taught artist. No teacher taught me to paint, write, act, or make music. Anything I learned from other people came much later, when I was already well into my forties and in conversation with people who held advanced degrees in philosophy, had built established careers and businesses in the arts, or belonged to an older generation and had already spent a lifetime mastering their disciplines. They were mentors, but the foundation had already been built through my own study, effort, failure, and persistence.

So I do believe that suffering is a component of being an artist. Full stop.

Whether we learn from a teacher or teach ourselves, part of the suffering is the disappointment of failure. Failure occurs constantly while we are learning. It occurs in every branch of art: medicine at its highest level, hands-on craftsmanship, performance, music, writing, and the visual arts. It always does.

We can debate the definition of suffering and how the word applies to art, but suffering is simply something that occurs within the artistic process. The artist must learn to remain at the desk while it is occurring.

AI will never suffer in this way, and a prompter will not suffer to that degree through prompting. A prompter does not ordinarily have to plow through an exhausting day, come home, turn on the computer, and think, My God, I am going to be here for the next four to twelve hours, while facing the likelihood that the work may still fail.

That is not what prompting requires.

Someone writes a prompt. You generate the image. Then the person says, “Look what I did.”

And that—far more than the artistic persistence people so often dismiss as vanity—is ego.

I am going to say something that will anger many people, but I believe it is true: most people who call themselves artists because they prompt an AI system to make images are driven primarily by ego.

It is a very different ego from the one artists are accused of possessing.

Yes, artists want to know that their work has moved or touched someone. But for the most part, we make the work for ourselves and for the world. And when I say we make it for the world, I mean that we make it for the people we believe may connect with it, gain something from it, or feel less alone because it exists.

Perhaps we want the viewer to learn something. Perhaps we want to communicate something political, philosophical, personal, or comic. Perhaps we are trying to reveal an injustice, preserve a memory, produce beauty, make someone laugh, or give shape to an experience another person has never been able to articulate.

Whatever the purpose, it remains motivated by desire. And a certain amount of suffering is involved because desire alone does not produce the work. The artist must acquire the knowledge, develop the ability, endure failure, and physically make the thing.

It is all extraordinarily human.

Sasha: Then I should not have allowed the word desire to displace suffering. Desire explains why the artist returns, but suffering describes much of what the artist must pass through in order to continue returning.

The distinction you are making is not that pain confers artistic legitimacy. It is that the practice requires a developed capacity to remain in contact with the work while pain, fatigue, frustration, uncertainty, and failure are also present.

The artist does not wait for the ideal mental condition.

The artist works while depressed. While grieving. While angry. While financially frightened. While physically tired. While disappointed by yesterday’s failure. While uncertain whether today’s attempt will be any better.