From Lines of Code to Canvas: The True Story of the First AI-Generated Image
When we think of artificial intelligence and art today, our minds instantly jump to hyper-realistic masterpieces generated in seconds by typing a few words into a modern text-to-image tool. But the history of machine-made art stretches back much further than most realize—long before neural networks, cloud computing, or modern GPUs.
The story of the very first AI-generated art begins not with a high-tech prompt, but with a visionary artist, a computer in the 1970s, and a pioneering program named AARON.
Meet Harold Cohen and AARON
In the late 1960s, artist Harold Cohen was a successful painter representing Britain at prestigious global art events like the Venice Biennale. However, he became fascinated by a radical question: What is the core essence of drawing? What are the absolute minimum cognitive rules required for a human brain to look at a collection of lines and recognize them as a figure or an object?
To answer this, Cohen took a sharp pivot. He moved to the United States and began teaching himself computer programming. By 1973, while working at the Stanford Artificial Intelligence Laboratory, he birthed a computer program that would change art history: AARON.
How the First “AI” Created Art
AARON was entirely different from today’s machine learning models. Modern generators are trained by scraping billions of images from the internet, looking for statistical patterns. AARON, by contrast, had zero access to a dataset or the internet.
Instead, Cohen manually wrote rules of art and geometry into the code. He taught the program:
- How objects overlap in space.
- The rules of perspective and structure.
- How to balance compositions.
Armed with these rules, AARON was able to make independent, autonomous decisions on how to compose a drawing. Connected to a specialized mechanical drawing machine (a flatbed plotter controlled by the computer), the program would grip a pen and physically ink abstract shapes, plant-like forms, and human figures onto paper.
The images produced by early AARON were simple, black-and-white line drawings. Yet, they represented a seismic shift: art created independently by a machine following logical rules rather than a human hand guiding a brush.
A Legacy That Echoes Today
Over the decades, Cohen continually updated AARON. As technology evolved, so did the program; AARON eventually learned to handle color, paint brushstrokes, and fill canvases with complex jungle scenes. Its works were eventually displayed in legendary institutions like the Tate Modern and the Whitney Museum of American Art.
Leave a Reply