What Was the First AI-Generated Image in History?

The Genesis of Artificial Creativity: What Was the First AI-Generated Image?

When we think of AI-generated images today, our minds instantly race to hyper-realistic photos, surreal digital paintings, and cinematic wonders conjured in seconds via text prompts like Midjourney, DALL-E, or Stable Diffusion. We live in an era where pixels are bent to our will by massive neural networks trained on billions of images.

Yet, the story of artificial intelligence making art didn’t begin in a Silicon Valley tech lab with deep learning algorithms. It began decades earlier, born from a surprising marriage of computer science, philosophy, and traditional fine art.

The undisputed pioneer of this movement is a computer program called AARON, and the first true AI-generated images date back to the early 1970s.

Meet Harold Cohen and AARON

To understand the first AI-generated image, you have to meet its creator: Harold Cohen, a celebrated British painter and professor. In the late 1960s, Cohen was already enjoying a successful traditional art career, representing Britain at prestigious global showcases like the Venice Biennale.

However, Cohen grew deeply curious about a fundamental question: What is the essence of drawing? What are the cognitive rules that allow a human brain to translate a mental concept into lines and shapes on a physical surface?

Instead of answering this philosophically, Cohen decided to approach it computationally. In 1968, he moved to the University of California, San Diego, and began teaching himself computer programming. By 1973, working extensively at the Stanford Artificial Intelligence Laboratory, he birthed AARON—not just a piece of software, but an autonomous artistic entity.

How the First AI “Painted”

AARON operated in a completely different universe than modern text-to-image generators.

• No Neural Networks: In the 1970s, “machine learning” as we know it didn’t exist. There was no scraping the internet for billions of JPEGs.
• Rule-Based Intelligence: Instead, AARON was built on “symbolic AI” or Good Old-Fashioned AI (GOFAI). Cohen laboriously codified his own understanding of art into the software.
• Autonomous Decision Making: Cohen didn’t dictate every stroke. He programmed rules regarding composition, overlapping objects, perspective, proportion, and closure. Within those boundaries, AARON made independent choices on how to layout and execute the drawing.

The earliest iterations of AARON produced stark, abstract, black-and-white line drawings. Connected to a custom drawing machine—a robotic arm holding a pen—the computer would physically move across a flatbed sheet of paper on the floor, plotting out strokes line by line while humans watched in sheer astonishment.

From Lines to Legacies

As computing power evolved over the decades, so did AARON.

• In the 1980s and 1990s, Cohen expanded the program’s capabilities, teaching it to handle color selection and complex organic shapes, eventually evolving to generate figures, rocks, plants, and intricate jungle scenes.
• AARON’s creations transcended the label of mere “computer experiments.” They were exhibited in world-class institutions like the Tate Modern in London and the Whitney Museum of American Art in New York.

Cohen famously viewed AARON as a collaborative partner rather than a mere tool. When Cohen passed away in 2016, AARON’s active development ceased, but its historical footprint remained massive.

Why the First AI Image Still Matters

Looking back, AARON’s minimalist 1973 line drawings look quaint next to today’s cinematic AI outputs. Yet, they represent a monumental psychological and technological shift.