This is Scientific American 60-Second Science. I'm Christopher Entagliata.

Artificial intelligence systems have bested humans at chess, poker, jeopardy, go, and countless

other games, but machines still aren't that great at understanding some basic rules about

the physical world.

They still can't do what three-month-olds do, and I think this is, you know, I'm a champion

of babies at the end of the day, and this is a clear win for babies.

These are still slam-dunking, even our most powerful computers when it comes to intuitive

physics.

Cognitive psychologist Susan Hesbos of Northwestern University listed off a few examples of those

intuitive physics principles, like solidity. Your coffee cup does not just fall right through

the table, or continuity. Objects don't just blink in and out of existence, and boundedness.

When you pick up your coffee cup, it sticks together. You don't end up with just the handle.

Babies know all three of these things, as early as three months of age. Their visual acuity

is lousy, the world is blurry, they could barely grasp this stuff. Babies get a lot of things

wrong, but it's just these initial kernels that get elaborated and refined through experience

in the world.

Now, computer engineers have taken a page from the Baby Playbook. Researchers at Deep Mind,

that's the AI company that trained computers to beat humans at Go. They've endowed a machine

learning system with certain kernels of knowledge about intuitive physics built in.

A kin to what an infant might be equipped with.

And after watching the equivalent of about 28 hours of training videos, showing things

like balls rolling and block-strapping, the AI system actually showed surprise when it

...