The Genesis of Artificial Creativity: The Story Behind the World’s First AI Image
Long before the internet was flooded with hyper-realistic, prompt-engineered wonders generated by tools like Midjourney or DALL-E, a quiet revolution was taking place in a university computer lab.
Today, generating an image takes mere seconds and a couple of typed sentences. But the history of artificial intelligence art didn’t begin with billions of parameters, massive neural networks, or scraped internet datasets. It began in 1973 with a British painter, a mainframe computer, and a software program named AARON.
Meet Harold Cohen and AARON
In the late 1960s, Harold Cohen was a successful, internationally acclaimed painter who had represented Great Britain at the prestigious Venice Biennale. Yet, he found himself gripped by an unorthodox curiosity: What is the core essence of drawing? Could a machine be taught to understand the cognitive steps a human artist takes to make sense of a blank canvas?
To find out, Cohen left the traditional art world behind, taught himself computer programming, and moved to the University of California, San Diego. By 1973, he had birthed AARON—the world’s earliest artificial intelligence art-generating system.
How the First AI Image Was Made
The way AARON operated was fundamentally different from modern AI. Today’s models function through machine learning, analyzing millions of preexisting photographs and paintings to spot statistical patterns.
AARON, conversely, had no access to a dataset of images. Instead, Cohen approached the computer like a human art student. He manually translated the cognitive rules of art into code:
- Geometry and Space: How lines relate to one another and intersect.
- Proportion and Composition: The rules governing balance, placement, and perspective.
- Physical Interaction: How objects overlap, occlude, and exist in a shared environment.
Crucially, Cohen didn’t draw the images himself using a computer as a remote-controlled pen. He wrote a system that possessed autonomous creative agency. Once running, AARON made its own real-time compositional choices. It decided where lines would go, what shapes would form, and how the layout would unfold—meaning no two outputs were ever identical.
From Primitive Lines to Gallery Walls
The earliest outputs of AARON were modest: stark, black-and-white abstract line drawings rendered physically onto paper by a custom-built robotic drawing machine connected to the computer.
To modern eyes, these primitive compositions might look simple, even rudimentary, compared to contemporary digital art. But in the 1970s, they represented a seismic philosophical shift. They proved that a machine could do more than crunch numbers; it could participate in the creative act.
As computer processing power evolved over the decades, so did AARON. Cohen continuously updated the program, eventually enabling it to understand color, shading, and complex figurative shapes like human figures, rocks, and lush jungle foliage. AARON’s work transitioned from computer laboratories into world-renowned spaces, including the Tate Modern and the Whitney Museum of American Art.
Why the First AI Image Still Matters
When Harold Cohen passed away in 2016, he left behind a profound legacy that bridges the gap between cold technology and human expression.
Leave a Reply