I think Go will eventually be conquered by technological advances that boggle our minds today.

I wonder – can we design a game that relies on intuitive leaps that can’t be outplayed someday by a brute force computer?

Great piece, lots to think about.

- In tic-tac-toe, the first player has an advantage, but nothing that the second player can’t neutralize.

- In chess, the first player has a significant advantage, and the second player generally doesn’t expect to win if the players are evenly matched. That’s why the players alternate between black and white in a match.

- In Go, the first player has a significant advantage, and the second player is compensated with a certain number of points added to their score. Usually 4.5 – 7.5, though there’s no consensus on what value would make the game balanced.

Computers are great for handling certain tasks, and I’m sure we’ll have them implanted in us someday. They make life better, and will continue to do so. But I’m also concerned about the day when they decide they don’t need us.

You are correct. In my head I was equating a draw game and a game where the first person always wins (or something similar) as the same thing. Obviously I failed to state that.

Well played

@Hunter:

This post reminds me of the song Kasparov vs. Deep Blue by Moxy Früvous. (warning, lots of pop-up ads (it’s a lyrics site, you know how those are) and some off-color language.)

I’m an artificial intelligence researcher by day (and superhero by night) so I’ve thought a lot about this sort of thing. As for games, my personal opinion is that I don’t care. If you’re playing a game to have fun, then have fun, regardless of whether the game is solved and regardless of what computers can and can’t do.

As to how this carries over to other types of problems, I think the trend of “advanced chess” is the way to go. Just like having easy access to Google and Wikipedia can augment an intelligent person’s ability to solve interesting problems, I think that as computers become able to solve simple common-sense reasoning problems, the most effective people will be those who delegate those simple problems to the computers and perform the high-level reasoning on their own. In your words, do the strategy and delegate the tactics.

Go is probably a draw game to, but as you pointed out the calculations are so large it can’t be done at this time. After reading this I would think any game with static rules is a draw game. Only a game with no rules and true randomness could actually be considered a non-draw game. War being one example (yes, it’s not a game in the spirit of games). However, a sufficiently complex draw game like chess or Go would be worth playing against other humans and computers…unless it’s chess against one of the best supercomputers.

I do like the idea of people and computers teaming up. Actually, I have always thought it was inevitable humans would team with computers and at some point actually have computers permanently implanted. Like your post says the computer could memorize stuff and do calculations freeing up the human for the strategy part of anything.

It would be a really powerful tool in research if ever researcher essentially knew every bit of information at all times, but was still able to make leaps of intuition.

We could actually teach people how to think in school instead of teaching them how to memorize. Yes, I know they try to teach people how to think now, but most teachers just fall back on memorization.