For centuries, a well-executed work spoke to the nature of its creator. A polished surface, a precise drawing, a tuned voice, or a clean print were the visible hallmarks of a craft: attaining such a standard required time, study, and trained hands. The finish functioned as a credential. Upon observing it, one inferred the skill behind it, and that skill constituted a portion of what was admired.
That connection is currently fracturing, and not for the first time. Today, an image with an impeccable finish can be produced in seconds, devoid of craft or even intention, simply by typing a phrase. When technical perfection is obtained in this manner, it ceases to prove what it once did. One can no longer deduce the labour sustaining a clean surface, for there may be none. The finish remains, yet it has forfeited its value as evidence. It no longer informs the observer of the nature of the attention, effort, or perspective contained within the work, as it may contain none at all.
I do not maintain that the imperfect is now of value, nor that clumsiness is a virtue. That would be to substitute one superstition for another: we once believed that a polished finish proved talent, and we would move on to believe that a crude one proves authenticity. Neither is the case. A defect also demonstrates nothing, as it may be manufactured at will. What is occurring is more radical and more simple: the finish, whether polished or broken, has ceased to inform. It is no longer a reliable indicator of anything.
None of this is new, and therein lies the interest. Each time a machine has made a previously rare skill more affordable, the same has occurred. When the Kodak camera arrived in 1888 with its slogan 'You press the button, we do the rest', anyone could obtain a sharp image of a scene, something that previously required mastery of chemistry, plates, and the darkroom. Sharpness ceased to distinguish. And the critic John Szarkowski, when he sought to explain what then made a photograph valuable, did not speak of perfection or defect: he spoke of selection, of framing, of the chosen instant, of the point of view. When the machine handles the execution, what remains on the human side is the decision: what to look at, when, from where, and what to leave out.
Before photography, and in a more brutal fashion, Duchamp stated it. When in 1917 he presented an industrial urinal as a work of art, he maintained something that a century later we have yet to fully assimilate: that the artist's hand was not the condition of art. He had not manufactured it; he had chosen it, displaced it, titled it, and compelled it to be viewed in a different manner. Manual skill, which seemed the heart of the matter, proved not to be. Artificial intelligence does not inaugurate this separation between knowing how to do and making art: it carries it to the extreme, it makes it mundane, but the rift has been open for over one hundred years.
What technical abundance alters is not, therefore, the definition of art, but rather what is scarce. When producing a well-finished image was difficult, that difficulty itself constituted a form of value, and at times it obscured other elements: a work could impress through its execution even if it conveyed little. Now that execution is effortless and ubiquitous, it has ceased to be a merit in its own right, and with it, the veil has fallen. What remains visible is what the finish sometimes concealed: whether the work selects something, whether it articulates a perspective, or whether it provokes a shift in the observer that was not present before. That remains rare, and now there is no polished surface to disguise or substitute for it.
There is a temptation to which one must not succumb: the belief that this is decided by technology alone, as if by law. No such law exists. When a skill becomes commodified, value does not always migrate to the same location: at times it moves to the viewpoint, at others to the concept, to the authenticity of the provenance, or to the relationship the work establishes with its audience. History does not possess a single, predetermined conclusion. The only certainty is what is lost: the alibi of the finish, the possibility of conflating being well-made with having something to say.
Perhaps this is less a threat and more a cleansing. For a long time, technical perfection could masquerade as artistic merit, and it frequently succeeded. Now that it is accessible to all, it deceives no one. The remaining question is more uncomfortable and more honest, and it is the perennial one, only now without refuge: once the sheen is removed, what does this work do that I cannot ignore? When the finish ceases to vouch for the work, the work must answer for itself.
On the open conversation
This text posits that when a technology renders common a skill that once demonstrated craft, that skill ceases to distinguish, yet imperfection does not take its place: what diminishes is the evidentiary value of the finish, not beauty. There is no singular destination for artistic value; what is laid bare is the question of what a work does before the observer, which stems from my research into the surplus. I accept verified cases as certain and distinguish the historical fact from the established narrative. Should anyone wish to intervene from the fields of painting, photography, musical production, art theory, or work with generative tools, this notebook remains open.
Sources
John Szarkowski, The Photographer’s Eye (The Museum of Modern Art, 1964 exhibition and 1966 book): significant photography as selection, framing, time, and point of view.
The Kodak No. 1 camera (1888) and its slogan 'You press the button, we do the rest' (National Science and Media Museum / Public Domain Review).
Marcel Duchamp, Fountain (1917) and the readymade; 'The Creative Act' (1957), on the participation of the spectator.
On nineteenth-century photography and painting: Tate Research (Clare A. P. Willsdon) and Cornell University Library.
Auto-Tune (Antares, 1997) and the debate on vocal correction and effect; history of drum machines and the fear of substitution (JSTOR Daily).
On the attention economy and scarcity: Herbert A. Simon, 'Designing Organizations for an Information-Rich World' (1971): 'a wealth of information creates a poverty of attention'.
On generative AI and value: Sungjin Park, 'The Work of Art in the Age of Generative AI' (AI & Society, 2024); Salas Espasa and Camacho, 'From Aura to Semi-Aura' (AI & Society, 2025).