Before the Prompt: The Story of AARON and the World’s First AI-Generated Image
Long before the era of typing text prompts into web browsers to summon hyper-realistic digital paintings, artificial intelligence art was born not in a tech startup, but in an artist’s studio.
In the late 1960s and early 1970s, while most computers filled entire air-conditioned rooms and processed data via punch cards, a British-born painter named Harold Cohen was asking a radical question: What is the minimum cognitive requirement for drawing?
His exploration led to the creation of AARON, widely recognized as the world’s first artificial intelligence art-making system—and a profound ancestor to the generative tools we use today.
The Mind of AARON
To understand the first AI-generated images, you have to forget everything you know about modern machine learning. Today’s models like DALL-E or Midjourney are trained on billions of images scraped from the internet, using statistical probability to stitch pixels together based on patterns.
AARON worked entirely differently. Developed initially while Cohen was a professor at the University of California, San Diego, and refined during a prestigious residency at the Stanford Artificial Intelligence Laboratory in the early 1970s, AARON was built on symbolic reasoning (often called Good Old-Fashioned AI, or GOFAI).
Cohen didn’t feed AARON a dataset of other people’s art. Instead, he manually coded the rules of the physical world and the visual arts into the program. He taught the machine:
- How objects occlude (overlap) one another.
- Basic principles of human anatomy and posture.
- Rules of composition, balance, and perspective.
In 1973, running on massive mainframe computers, AARON produced its earliest breakthrough drawings.
What Did the First AI Images Look Like?
The earliest outputs of AARON were not explosions of color or photorealistic portraits. They were stark, elegant, black-and-white line drawings.
Using custom-built robotic drawing machines—often referred to as “turtles”—connected to the computer, AARON would autonomously plot coordinates on large sheets of paper, dragging a pen across the surface. The resulting images featured abstract organic shapes, primitive human figures, and botanical motifs.
Crucially, Cohen did not pre-design the final composition. While he wrote the rules, AARON made autonomous decisions on the fly about how to arrange elements, balance negative space, and complete the drawing. In a true philosophical sense, the machine was executing creative agency within the boundaries of its programmed universe.
A Legacy Carved in Ink
As computing power evolved over the decades, so did AARON. By the 1980s and 1990s, Cohen had taught the program how to handle color, eventually moving from pen-and-paper drawings to massive digital projections and complex jungle scenes. AARON’s works were eventually exhibited in world-renowned institutions like the Tate Modern in London and the Whitney Museum of American Art in New York.
Cohen continued developing AARON until his passing in 2016, spending nearly 50 years with his digital collaborator.
Leave a Reply