In the mid-1990s, a supercomputer built by IBM called Deep Blue played chess against then world champion, Gary Kasparov.

There were two matches, each encompassing six games.

The first match was won by Kasparov four to two, but that loss netted Deep Blue, a world first,

the first computer program to defeat a world champion in a classical game of chess

under official tournament rules. The second match, though, which took place in mid-1997, went to Deep Blue,

3.5 to 2.5, those halves the result of three draws out of the six games. This earned deep blue the additional accolade

of having been the first computer program to defeat a world champion in a match, a complete

collection of six games, ever. This was, and still is by many, considered to be a milestone

moment in the history of the game of chess,

but also more broadly in the development of software,

of what you might call a type of artificial intelligence,

as long as we're sticking with a non-general intelligence definition of the term,

software that can do impressive computation and something that seems a bit like thinking, but which is really mostly just brute forcing problems with incredible algorithmic sophistication.

That's not meant to underplay the significance of the event, however.

Up until this point, it was considered a bit of a truism that chess was a game in which

humans would probably always maintain the upper hand

over our machines. Computers were powerful, but somewhat linear thinkers, after all. And chess required

creativity. The sheer number of possible moves and outcomes requires a sort of dexterity to one's

thinking. Having shown that this wasn't

necessarily the case, the dominant narrative shifted to the traditional Chinese board game,

Go, which has in some ways simpler rules than chess, but a far larger set of potential moves,

and thus it should require more human intuition.

...