There’s an old saying:“You are the average of the five people you spend the most time with.”

I want AI to do my laundry and dishes so that I can think, write, and create art, not for AI to do my art and writing so that I can do my laundry and dishes.

— Joanna Maciejewska

Joanna captures Moravec’s Paradox better than most textbooks.

Who is Hans Moravec?

Hans Moravec is a professor at Carnegie Mellon University, a futurist, and a philosopher.

In his 1988 book Mind Children, he articulated an idea that later came to be known as Moravec’s Paradox (the term itself was coined afterwards).

What is Moravec’s Paradox?

Moravec observed something deeply counterintuitive:

“It is comparatively easy to make computers exhibit adult-level performance on intelligence tests or playing checkers, and difficult or impossible to give them the skills of a one-year-old when it comes to perception and mobility.”

Essentially, AI finds the "hard" things easy and the "easy" things nearly impossible.

The "Easy" for AI:
Defeating Grandmasters at Chess or Go, solving complex calculus, and analysing massive datasets.

The "Hard" for AI:
Walking through a crowded Times Square without bumping into anyone, folding a pile of soft towels, or holding a conversation in a noisy bar (humans manage effortlessly; machines fail the auditory Turing test)

Why Is the "Simple" So Difficult?

Moravec argues that this dichotomy is a result of evolution.

For nearly a billion years, multicellular life has survived by refining sensory and motor skills. We have "encoded" a billion years of experience into our brains regarding how to navigate space, recognise faces, and react to danger. This knowledge is so deep and unconscious that we make the "difficult" look easy. We are all "prodigious Olympians" at simply existing in the physical world.

By contrast, abstract thought (math, logic, and chess) is a "new trick" in evolutionary terms, less than 100,000 years old. We haven't mastered it yet, so it feels hard to us. But for a machine, logic is just a set of instructions. It doesn't require a billion years of "street smarts" to solve an equation.

The real paradox:

Intelligence we assumed to be uniquely human and impossibly complex turns out to be teachable to machines.

Meanwhile, riding a bicycle, something child learns in a few hours, remains astonishingly hard to teach robots.

But nature, as always, has a wicked sense of humour, quietly reminding us that our narcissism may be the least intelligent thing about us.

Take it easy until next time.

Blogging is something I enjoy, and I share my thoughts on my blog most weekends. Explore all my blogs at https://lnkd.in/ejq7CWaQ

Views are my own.

Blogging is something I enjoy, and I share my thoughts on my blog most weekends.

Read all my “Notes to Self” at view all blogs.