Can Artificial Intelligence Create Haitian Art?

What Artificial Intelligence Cannot Replicate About Culture, Memory and the Human Imagination

An Essay from the Private Collection of Myriam Nader Salomon

By Myriam Nader Salomon, Haitian Art Dealer, Appraiser and Founder of Myriam Nader Haitian Art Gallery, NY

I have been watching artificial intelligence move through the art world with a mix of curiosity and skepticism. It can generate a finished image in seconds, and for a long time I could not quite make sense of how that was possible. What finally pushed me to put these thoughts into writing was reading Pope Leo XIV's 2026 encyclical Magnifica Humanitas, on safeguarding the human person in the age of artificial intelligence. Its consideration of both the promise and the limits of AI gave language to questions I had already been asking through my work with Haitian art.

Ask an AI image generator for a colorful Haitian market scene, a tropical landscape, a Vodou-inspired composition, or even an image that resembles the work of a particular Haitian artist, and the result can be surprisingly convincing. AI is now being used in art classes as a teaching and reference tool.

But convincing is not the same as authentic. What AI produces is a recombination of patterns it has already seen. It does not walk through a market in Jacmel. It does not carry the weight of Haitian history, the rhythm of a Vodou ceremony, or the memory of a childhood spent among these images and sounds. It cannot replicate lived experience, only imitate its surface.

This is the distinction I keep returning to as I write about this subject. A Haitian painting made by hand carries the culture, memory, and imagination of its maker. That is not a technical detail an algorithm can reverse engineer. It is the entire reason the work matters.

But is it Haitian art?

That question extends far beyond technology. It touches on authorship, culture, history, spirituality, authenticity, memory, and ultimately on what gives a work of art meaning.

As artificial intelligence becomes increasingly capable of imitating recognizable artistic styles, collectors, artists, scholars, galleries, and museums may need to draw a more important distinction:

Reproducing the appearance of a culture is not the same as creating from within that culture.

Jacques-Enguerrand Gourgue, The Forbidden Fruit, 1994. Private Collection of Myriam Nader. Gourgue's intensely personal visual language offers a compelling point of departure for considering human imagination, knowledge, culture, and creativity in the age of artificial intelligence.

AI Can Learn a Style, But Can It Live a Culture?

Artificial intelligence learns by identifying patterns in vast amounts of data. When applied to images, it can recognize recurring colors, forms, subjects, compositions, textures, and stylistic characteristics, and use those patterns to generate new images.

Applied to Haitian art, an AI system might recognize crowded market scenes, lush vegetation, Vodou imagery, rural landscapes, historical subjects, intricate patterning, brilliant colors, flattened perspectives, or particular approaches to depicting people and nature.

It can learn what these things look like.

What it cannot do is experience what they mean.

Haitian art emerged from an extraordinary and complex history shaped by African heritage, Indigenous and European influences, colonialism, revolution, independence, Catholicism, Vodou, rural traditions, urban life, political upheaval, migration, spirituality, humor, resilience, memory, and imagination.

These are not simply visual ingredients. They are lived experiences.

And lived experience cannot be downloaded into an algorithm.

Wilson Bigaud's depiction of a Vodou ceremony offers a powerful example of the distinction between visual recognition and lived cultural experience. An artificial intelligence system may identify musicians, dancers, ritual activity, gestures, clothing, and other visual elements, but recognizing those elements is not equivalent to understanding the spiritual, cultural, and communal experience the artist represents.

Cultural Experience Is Not the Same as Machine Recognition.

The distinction becomes especially compelling in Wilson Bigaud's depiction of a Vodou ceremony. An artificial intelligence system can analyze the visible elements of a scene. It may identify figures, musicians, movement, clothing, gestures, objects, colors, and compositional relationships. With sufficient training data, it may even recognize that the subject relates to Haitian Vodou.

But recognition is not experience. Vodou cannot be reduced to a set of visual symbols that an algorithm can identify and recombine. It is a living religious and cultural tradition with histories, beliefs, rituals, communities, spiritual meanings, and forms of knowledge that extend beyond what is visible in a painting.

Bigaud was interpreting a world embedded in Haitian life and culture. His painting therefore represents more than a scene composed of identifiable objects and figures. It reflects a human artist's observation and interpretation of a culturally meaningful experience. An AI system may learn to reproduce some of the visual vocabulary associated with such imagery, but it cannot have participated in the culture from which that vocabulary emerged.

That distinction goes to the heart of the question posed by this essay: imitating the appearance of a culture is not the same as creating from within it.

There Is No Single “Haitian Style”

One of the greatest dangers of generative AI is its tendency to transform complex cultural traditions into recognizable aesthetic formulas.

A prompt requesting “Haitian-style art” might produce tropical colors, markets, mountains, Vodou symbols, vegetation, village scenes, animals, or elaborate patterns.

But Haitian art is not one style.

Consider the extraordinary differences among Jacques-Enguerrand Gourgue, Prospère Pierre-Louis, Emmanuel Pierrette, Wilson Bigaud, Serge Moléon Blaise, Jean-Baptiste Bottex, Seymour Bottex, and Albott Bonhomme.

All are Haitian artists. Yet their visual languages are profoundly different.

Prospère Pierre-Louis, Three Loas, 1988. Pierre-Louis's distinctive visual vocabulary shows that Haitian art cannot be reduced to a single aesthetic formula or a generalized “Haitian style.”

The visionary language associated with Prospère Pierre-Louis differs radically from Emmanuel Pierrette's carefully observed scenes of everyday life.

Jacques-Enguerrand Gourgue's psychological and symbolic universe bears little resemblance to the lush imagined landscapes of Albott Bonhomme.

Wilson Bigaud's narrative figuration differs again.

These differences matter.

When we reduce Haitian art to a set of recognizable visual characteristics, we risk turning culture into decoration.

Style Is Not Authorship.

Artificial intelligence also forces us to reconsider something that specialists and serious collectors have always understood:

In my 37 years of experience, I have learned that a painting is not authentic simply because it looks like an artist's work.

Connoisseurship requires far more than superficial resemblance.

When evaluating a painting attributed to an important Haitian artist, specialists may consider composition, paint handling, materials, signature, inscriptions, subject matter, provenance, age, canvas or support, known periods of production, documentary evidence, and comparisons with authenticated examples.

Two paintings may appear remarkably similar yet have entirely different histories and values.

AI makes this distinction even more important because sophisticated visual imitation is becoming easier.

For collectors, the question can no longer be merely: “Does this look right?” It must also be: “Who created this object? When was it made? Where has it been? And what evidence supports that history?”

In my work with Haitian art, I have learned that attribution rarely hinges on a single characteristic. A signature may be important, but so are the materials, provenance, paint handling, composition, subject matter, and the many subtle characteristics that become recognizable through years of studying an artist's work.

The Physical Artwork Still Matters

There is something else an AI-generated image cannot reproduce: the historical life embedded in an original physical object.

A painting bears witness to its creation.

The surface may reveal brushwork, layering, corrections, aged pigments, craquelure, canvas characteristics, stretcher construction, inscriptions, labels, framing history, restoration, or other physical evidence. These characteristics are not imperfections to be eliminated. They are part of the object's biography.

For collectors, provenance is increasingly important. Who owned the painting? Where was it acquired? Was it purchased directly from the artist? Did it pass through a gallery, a private collection, an exhibition, an auction, or an estate?

Are there photographs, invoices, correspondence, certificates, exhibition records, or prior appraisals that document its history?

In a world where convincing images can be generated almost instantly, the physical and documentary history of an artwork may become more important, not less.

Art Can Carry Historical Memory

Haitian art also serves another function that cannot be understood solely through visual analysis.

It preserves memory.

Artists have repeatedly interpreted Haiti's history, revolution, religion, landscape, political life, and collective experience.

Serge Moléon Blaise, La Grande Bataille entre les Haïtiens et les Anglais en 1789, 1993. Historical painting can function not simply as illustration but as an artist's interpretation of collective memory and national history.

An AI system could be instructed to generate an imaginary historical battle scene. But an artist interpreting the history of his own country engages in something different.

The artist becomes part of the ongoing process by which a society remembers, interprets, and visualizes its past.

The distinction is subtle yet essential.

Information about history is not the same as historical memory.

Everyday Life Is Also Cultural Memory

History does not exist only in battles, revolutions, and political events. It also survives in ordinary life. Streets, houses, markets, clothing, transportation, religious observances, celebrations, conversations, and routines can become historical evidence when captured through an artist's eyes.

Emmanuel Pierrette, To The Market, 1994. Scenes of everyday life preserve details of place, architecture, social interaction, and lived experience that form part of Haiti's cultural memory.

A painting such as Emmanuel Pierrette's To The Market may initially appear to document an ordinary street scene. Yet over time, the ordinary becomes historical. Architecture changes. Clothing changes. Transportation changes. Neighborhoods change. Social customs change.

The artist has preserved a moment of human life. An algorithm can synthesize visual information about that moment. The original painting was created in time.

That distinction matters.

What Happens to Creativity When Machines Can Imitate It?

Artificial intelligence forces us to reconsider what we mean by the word “creative.”

If a machine produces a beautiful image, does beauty alone make it art?

Perhaps that is not the most interesting question. The deeper question is whether creativity is only about producing an outcome.

Human beings create for reasons that cannot always be explained.

An artist may work from memory, spirituality, observation, trauma, humor, longing, political conviction, experimentation, faith, imagination, or simply an inexplicable need to create something that did not previously exist. The finished artwork is therefore more than an arrangement of shapes and colors.

It is evidence that a particular human being existed, observed, imagined, and made choices.

Nature Is Not Simply Something an Artist Copies

Even seemingly straightforward landscape or animal paintings reveal the distinction between representation and human interpretation.

Three Birds and Cascades, 1994. Albott Bonhomme transforms Haiti's tropical environment into an idealized world of extraordinary vegetation, birds, water, atmosphere, and color. The painting reflects not simply nature observed, but nature reconstructed through an individual imagination.

An AI can generate thousands of imaginary tropical landscapes. But the significance of a painting does not depend merely upon whether the landscape depicted actually existed. Artists transform observation through memory and imagination.

They exaggerate. They simplify. They invent. They repeat certain motifs. They develop preferences.

Over time, these decisions become an artistic language.

That language belongs to an individual.

Why AI May Ultimately Make Original Art More Important

There is an intriguing possibility hidden within the anxiety about artificial intelligence. AI may ultimately remind us why original art is important.

When unlimited images can be generated almost instantaneously, scarcity changes meaning.

A physical painting created decades ago by a Haitian artist cannot be simply reproduced.

It belongs to a particular moment. Its materials have aged over time. Its brushstrokes record physical decisions. Its surface bears evidence of handling and history. Its provenance connects people, places, galleries, collectors, families, and sometimes generations.

Its imperfections are part of it.

AI can create another image, but it cannot recreate that object's past. The more effortless digital image production becomes, the more meaningful genuine evidence of human presence may be.

What Haitian Art Collectors Should Consider in the Age of AI?

Collectors should increasingly focus on authorship, provenance, documentation, condition, and specialist knowledge, especially when acquiring important Haitian works.

A compelling image is no longer enough.

Before acquiring a significant work, collectors should consider the artist’s identity, whether the attribution is supported, what is known about the artwork's provenance, whether the materials and construction are consistent with the purported period, how the painting compares with documented examples, whether signatures or inscriptions are consistent, and whether supporting documentation survives.

Artificial intelligence may become an extraordinary research tool. It can help organize information, compare images, translate documents, identify research questions, and detect patterns that warrant further investigation.

But it should not replace connoisseurship, provenance research, physical examination, documentary evidence, or informed professional judgment.

Can AI Create Haitian Art?

AI can certainly generate images that look Haitian. It can generate a marketplace. It can create mountains and tropical vegetation. It can imitate certain compositional structures. It can reproduce recognizable symbols. It may even generate an image that, at first glance, appears reminiscent of a known Haitian artist’s work.

But appearance is only one aspect of art.

Haitian art is inseparable from the people who created it and from the historical and cultural worlds from which those artists emerged.

Behind an authentic work is an artist. Behind the artist is experience. Behind that experience lies history.

Behind history lie generations of human beings remembering, believing, struggling, celebrating, imagining, and creating.

Artificial intelligence can analyze the visual evidence from those experiences. It cannot have lived them.

The Future of Haitian Art Is Still Human

Artificial intelligence will undoubtedly shape the future of art.

Artists may embrace, reject, challenge, criticize, or incorporate it into forms of creative expression we have not yet imagined.

That evolution does not diminish the importance of traditional art.

It may help us understand it more deeply.

Perhaps the most important question raised by artificial intelligence is not: “Can a machine make art?” It is: “What makes human art human?”

Artificial intelligence is not human intelligence. It can process, calculate, and generate. It cannot remember, believe, or create from lived experience. That distinction is the one Haitian art keeps asking us to protect.

For Haitian art, the answer lies not merely in color, composition, subject matter, or style.

It exists in memory, in culture, in spirituality, in history, and in individual imagination.

And in the physical evidence of one human being leaving something behind for another human being to encounter. That is something no prompt can reproduce.

Thirty-seven years in this field have not dulled that fascination. If anything, the longer I study Haitian paintings, the more I notice how much of the artist remains present in the work. Every brushstroke is a trace of a hand, a choice made at one moment, in one place, under conditions that will never recur.

I do not dismiss what technology offers to the public. High-resolution imaging reveals brushwork and layering that I could not have examined this closely a generation ago, and digital archives make it far easier to trace a painting's history against known works. I carefully rely on these tools regularly.

But none of it captures what truly lives within an original piece. A painting made in Haiti carries the place it came from, the imagination of the person who made it, and a human experience no algorithm can reconstruct afterward. A machine can study a pattern. It was not there when that pattern was formed.

That is why authentication and provenance carry so much weight in my work. Confirming a signature is only part of it. The larger task is honoring the fact that a real person, shaped by a real culture and history, created something that had never existed before. That is the part technology cannot touch, and it is the part worth protecting.

In Summary

My sense is that AI can feel scary when it moves this fast, and that reaction is understandable. My take isn't that it's dangerous, but that it has real limitations worth naming. It can imitate patterns beautifully, but it cannot live a culture, carry memory, or replicate the lived experience behind a Haitian painting. That distinction is what I wanted to highlight in this piece.

Style is not authorship, and a convincing image is not the same as an authentic one. As generated imagery becomes effortless, the human hand, the physical object, and documented provenance matter more, not less. If anything, AI may increase the importance collectors place on authentic original Haitian artwork, as genuine evidence of human presence grows rarer and more meaningful.

AI will continue to change the art world, for better and for worse. But it also reminds us why the original, human-made work matters even more. It can support collectors, dealers, and appraisers as a research tool, but it cannot replace connoisseurship or the fact that a real artist, shaped by a real history, created something that never existed before.

About the Collection

The artworks illustrated in this essay are drawn from my private collection, reflecting the remarkable diversity of artistic expression within Haitian art.

Explore the Private Collection of Myriam Nader Salomon