The True Story of the First AI-Generated Image in 1973

Beyond the Algorithm: How 1973 Sparked the Era of AI Art

When people think of artificial intelligence generating images, they usually picture modern tools that whip up hyper-realistic photos or fantastical digital paintings in mere seconds. But the roots of AI-generated art stretch back much further than most realize.

Long before neural networks and diffusion models dominated the tech landscape, the groundwork for machine-made art was laid in 1973 by a visionary artist who dared to ask what would happen if a computer could learn how to draw.

Meet AARON: The World’s First AI Artist

In 1973, British-born painter Harold Cohen—then a professor at the University of California, San Diego—began working on a revolutionary computer program that would eventually be named AARON.

At a time when computers were massive, room-sized machines with minimal computing power, Cohen wasn’t trying to use the computer as a simple digital paintbrush or a tool to trace pre-existing designs. Instead, he wanted to build a collaborator. He wanted to encode human cognitive rules about drawing, perspective, and the physical world into a machine, allowing the computer to make its own creative decisions.

“Cohen seeded Aaron with all sorts of knowledge: about basic objects, physics and fundamental techniques of drawing.” — Smithsonian Magazine

How It Worked in the 1970s

Unlike today’s text-to-image generators that rely on massive datasets scraped from the internet, AARON operated on symbolic reasoning and rule-based programming.

• The Rules of Reality: Cohen didn’t teach AARON by showing it millions of photos. Instead, he wrote code that translated fundamental rules of art and geometry. He taught the program how objects overlap, how lines form shapes, and how to compose elements on a page.
• The Robotic Arm: In its earliest iterations, AARON’s digital outputs were translated into physical reality using custom-built drawing machines—often flatbed plotters or robotic arms holding pens and brushes. Cohen would watch as the machine autonomously maneuvered across paper, executing lines and abstract shapes it had computed on its own.

The Philosophy Behind the Pixels (and Ink)

AARON’s creation triggered deep questions that artists and computer scientists are still debating today: Who is the actual artist? Is it the human who wrote the code, or the machine that executed the creative choices?

Cohen famously viewed his relationship with AARON as a partnership, akin to a master artist working with a studio assistant. Over the decades, as computer hardware evolved, Cohen continuously updated AARON. What started as abstract, geometric line drawings in the 1970s eventually evolved into complex botanical structures, figures, and vibrant, colorful jungle scenes by the 2000s—all driven by the core intelligence Cohen first sparked in 1973.

Why 1973 Matters Today

Today’s generative AI boom owes a massive debt of gratitude to pioneers like Harold Cohen. Long before deep learning and graphics cards revolutionized the field, AARON proved that art and code could merge. It shattered the notion that computers were strictly mathematical tools incapable of creative output.