The Birth of Machine Imagination: How 1973 Gave Us the First AI-Generated Images
Long before modern graphics cards hummed in server racks, text prompts conjured hyper-realistic portraits, and internet datasets fueled high-tech models, the genesis of artificial intelligence art took root.
In 1973, an acclaimed British painter did something radical. Instead of using a brush on canvas, he sat down at a computer terminal and set a machine loose to draw entirely on its own. His name was Harold Cohen, and his creation, an autonomous software program called AARON, permanently changed the trajectory of visual culture.
From the Canvas to the Codebase
By the early 1970s, Harold Cohen was already an established name in the international art world, having proudly represented Great Britain at the prestigious Venice Biennale. Yet, he grew intensely curious about the cognitive mechanics behind human art. What actually constitutes a line? How do humans instinctively know when a shape is “closed” or how figures relate to a background?
Rather than treating the computer as a mere digital paintbrush or a glorified copy machine, Cohen wanted to understand if artistic decision-making could be formally codified.
He packed up his studio and took up a fellowship at the Stanford Artificial Intelligence Laboratory (and later the University of California, San Diego), diving headfirst into computer science. At a time when researchers were primarily teaching early computers to solve logic puzzles and play rudimentary chess, Cohen attempted something entirely unprecedented: teaching a machine to think like an artist.
Meet AARON: The World’s First AI Artist
In 1973, Cohen’s deep programming efforts culminated in the unveiling of AARON.
Unlike today’s machine learning models, which ingest billions of images to recognize statistical patterns, AARON was built on symbolic AI and rule-based logic. Cohen meticulously translated his own understanding of cognitive drawing rules into software. He wrote instructions governing:
- How lines could intersect and bound space.
- How shapes could be nested or balanced across a composition.
- When a region or drawing was formally “complete”.
Once Cohen pressed “start,” he did not intervene. AARON would autonomously make its own creative choices within those programmed boundaries, sending physical coordinates to a connected mechanical drawing table (a pen plotter) that dragged an ink pen across paper.
The resulting 1973 outputs were intricate, abstract compositions of lines and closed forms. Crucially, no two drawings were identical, and none required human guidance during execution.
Why the 1973 Milestone Still Matters Today
“AARON demonstrated that symbolic AI could encode aesthetic judgement well enough to produce coherent, original visual work without human intervention at the point of creation.” — Tech & Art Historians
The creations of 1973 spark a vital philosophical realization: AI art is not a 21st-century invention.
While the aesthetic fidelity of AARON’s early black-and-white line drawings looks vastly different from modern generative models, the core concept remains identical. Cohen proved that creativity was not an exclusively mystical human trait impervious to replication, but a process capable of being understood, modeled, and computed.
Leave a Reply