Why Artificial General Intelligence (AGI) Is So Hard To Get To

The challenge of creating a machine or computer matching human intelligence across the board (AGI) is one of the most demanding tasks we face.

That is how difficult it is to understand our own brain and attempt to make something as complex and powerful as it is.

What’s curious is that the things we struggle with (algebra, language translations, beating a grandmaster at chess) are the things a computer can now excel at.

But the things we find simple to do (identifying a cat from a dog, walking around on two legs) are the things a computer struggles with.

This is Moravec’s paradox:

“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.”

Moravec, Hans (1988), Mind Children, Harvard University Press, p. 15

One possible explanation of this paradox offered by Moravec is based on evolution and the process of natural selection over hundreds of millions of years.

He believes evolution has allowed humans to perform what are largely unconscious and effortless skills such as perception, visualization, and motor skills to a degree that’s difficult for a computer to replicate.

“Encoded in the large, highly evolved sensory and motor portions of the human brain is a billion years of experience about the nature of the world and how to survive in it. The deliberate process we call reasoning is, I believe, the thinnest veneer of human thought, effective only because it is supported by this much older and much more powerful, though usually unconscious, sensorimotor knowledge. We are all prodigious Olympians in perceptual and motor areas, so good that we make the difficult look easy.”

Moravec, Hans (1988), Mind Children, Harvard University Press, p. 15-16

And this also includes social skills such as interpreting subtle facial expressions and understanding emotions, which humans have evolved to develop.

Whereas skills which are relatively new to humans (mathematics, engineering, scientific reasoning) a computer can beat us at because we haven’t had the time to evolve to a high level of mastery at them.