"Games (Chess, Go, Dota) represent closed systems, which means we humans filled the machine with a target, with rules. There is no automatic transfer of the knowledge that machines could accumulate in closed systems to open-ended systems." - Garry Kasparov

CONNECT:
- If you enjoyed this video, please subscribe to this channel.
- AI Podcast: https://lexfridman.com/ai/
- Show your support: https://www.patreon.com/lexfridman
- LinkedIn: https://www.linkedin.com/in/lexfridman
- Twitter: https://twitter.com/lexfridman
- Facebook: https://www.facebook.com/lexfridman
- Instagram: https://www.instagram.com/lexfridman
- Slack: https://deep-mit-slack.herokuapp.com

People who just look at my matches with Deep Blue 20 years ago or recent games are played by newer machines in Go or in popular video game Dota.

That makes a very important point that in all these cases, all these games are different.

Go, for instance, is much more complex than chess.

It's more abstract, more strategic, so it's more difficult to accumulate sophisticated knowledge as we did in chess.

But all these games, they represent closed systems, which means we humans fill the machine with a target, with rules.

There's no automatic transfer of the knowledge that machines could accumulate in closed systems to open-ended systems.

Machines can do many things, and they will learn even more things in the future, but I don't see any chance in the foreseeable future for machines to ask the right questions.

Now, they can ask questions, but they don't know what to ask.


Speaker 0: People who just look at my matches with Deep Blue 20 years ago or recent games are played by newer machines in Go or in popular video game Dota. That makes a very important point that in all these cases, all these games are different. Go, for instance, is much more complex than chess. It's more abstract, more strategic, so it's more difficult to accumulate sophisticated knowledge as we did in chess. But all these games, they represent closed systems, which means we humans fill the machine with a target, with rules.

There's no automatic transfer of the knowledge that machines could accumulate in closed systems to open-ended systems. Machines can do many things, and they will learn even more things in the future, but I don't see any chance in the foreseeable future for machines to ask the right questions. Now, they can ask questions, but they don't know what to ask.

See all Lex Fridman transcripts on Youtube

Garry Kasparov: Open-Ended AI