Long before the era of text-to-image prompts, neural networks, and hyper-realistic digital canvases, the history of artificial intelligence art began with a computer program named AARON.
Created in the late 1960s and early 1970s by British artist and programmer Harold Cohen, AARON is widely recognized as the world’s first AI-generated art system. Its very first autonomous drawings—produced around 1973 and 1974—marked a monumental turning point, proving that a machine could do more than crunch numbers; it could make creative decisions.
The Mind of AARON: Code Over Data
Unlike modern generative AI tools like Midjourney or Stable Diffusion, which devour billions of images to learn statistical patterns, AARON did not use machine learning or deep neural networks. Instead, it was a rule-based expert system.
Cohen spent decades meticulously translating his own understanding of human artistic logic into computer code. He did not program specific final images; rather, he programmed the rules of art. AARON was given instructions on:
- How lines intersect and relate to one another.
- The geometry of human figures, plants, and basic shapes.
- Principles of composition, spatial organization, and proportion.
Armed with these cognitive rules, the software was left to improvise. When running, AARON would make autonomous choices about what to draw next on a blank canvas without human intervention.
Bringing the Art to Life
Because digital screens weren’t capable of displaying high-resolution art in the early 1970s, AARON was connected to a physical plotting device—often referred to as a “turtle”—that held a pen and physically moved across a sheet of paper laid out on the floor or a table. Spectators could watch in real-time as the computer directed the mechanical arm to ink intricate, abstract, and organic black-and-white line drawings.
As computing power and programming techniques advanced over the following decades, Cohen continuously updated AARON. The software evolved from generating simple geometric line drawings into creating complex, colorful scenes featuring human figures, lush botanical environments, and rich textures. AARON’s work eventually transcended the laboratory, earning exhibitions at prestigious institutions like the Tate Modern and the Whitney Museum of American Art.
A Legacy That Paved the Way
Harold Cohen’s pioneering work with AARON laid the conceptual groundwork for everything that followed. While the technology shifted dramatically decades later—moving from Cohen’s hand-coded rules to Markov random fields, Variational Autoencoders (VAEs), Generative Adversarial Networks (GANs) in 2014, and modern diffusion models—the core philosophical question remained the same: Can a machine be an artist?
AARON proved that the answer was yes, transforming the computer from a passive calculator into an active creative partner long before the rest of the world imagined it possible.
Leave a Reply