Games teach things and/or make statements, whether they mean to or not. If you have 51 minutes of spare time, you can see Carnegie Mellon's Luis von Ahn talk about games he designed to specifically solve problems. He's interested in using "human computing cycles" (time that people "waste" playing games) to solve problems (such as labeling objects in images) that computers are not yet good at solving.
I hope this video is somewhat embarrassing to the entire game industry, as it took someone outside of the usual sphere of game-making to demonstrate the extreme power of games, and that they can be used for useful purposes. The games presented in the video are simple (and probably fun), but it's more about the radical shift in thinking about what a game could be used for that is notable.
(Thanks to Eric Williams for telling me about this in the first place, even though I'm now 8 months late to the party. Or three years late, if we count the release of the first game mentioned in the talk.)
This whole thing reminds me of a lecture given by 42 Entertainment, makers of the "i love bees" alternate reality game that surrounded the release of Halo 2 and "The Beast" AR game surrounding the release of the movie A.I.. Anyway, they also talked about how they could give out extremely difficult puzzles/problems, all of which would be solved nearly instantly by the hive mind of the world. They even started giving out problems to which they did not know how to find solutions, and even these were solved by the vast network of connected minds, organized across various chat rooms and forums.
Someone from 42 joked, "If we made a game that focused all this brainpower on solving world hunger, we'd have it solved in two days, tops."
Everyone laughed.
Except me.
--Sirlin