What Is the AI Effect? Definition & Examples

Moving Goalposts: Understanding the “AI Effect”

If you ask someone in the 1960s what artificial intelligence looks like, they might imagine a mechanical brain capable of logic, chess, and complex calculation. If you ask someone today, they might point to autonomous agents, generative text models, or systems that create photorealistic imagery. Yet, tucked inside this relentless march of progress is a strange psychological quirk of human perception: the AI effect.

Coined by computer historians and popularized by pioneers like John McCarthy, the AI effect describes the tendency of people to disqualify a technology from the “artificial intelligence” category the moment it actually works. As McCarthy once famously lamented, “As soon as it works, no one calls it AI anymore.”

The Shifting Definition of Intelligence

The phenomenon is anchored by a famous axiom often attributed to computer scientist Larry Tesler: “AI is whatever hasn’t been done yet.” Because artificial intelligence lacks a rigid, universally accepted boundary, the goalposts are perpetually moved by success.

Consider the history of computing milestones:

• Chess and Strategy: For decades, beating a human grandmaster at chess was treated as the ultimate litmus test for machine cognition. When IBM’s Deep Blue defeated Garry Kasparov in 1997, the reaction from some intellectuals wasn’t wonder at machine brilliance, but a reassessment of chess itself: I guess chess doesn’t require real thought after all.
• Everyday Utilities: Optical character recognition (OCR) used to read postal codes, email spam filters, speech-to-text transcription, and GPS route-finding were all once groundbreaking subfields of AI research. Today, they are treated as mundane background infrastructure. Once you can use it effortlessly on a smartphone, it ceases to be “intelligence” and simply becomes software.

Why the Mystery Matters

This cycle of de-classification is not mere ingratitude; it is rooted in how human beings process comprehension and mystery.

As long as a technology is opaque, mysterious, and out of reach, it feels magical—it feels like a mind. But the moment scientists crack the mechanism, write the algorithms, and make the system predictable, the illusion evaporates. Demystification kills wonder. We look under the hood, see statistical weights, linear algebra, and logic gates, and conclude: “Oh, that’s just math. That’s not real thinking.”

This creates a peculiar paradox for researchers and engineers. They spend years chasing a capability, only to watch the finish line sprint further away the second they cross it. The technologies we celebrate today as miraculous cognitive breakthroughs will inevitably become the invisible, taken-for-granted utilities of tomorrow.