I always like to compare holy grail problems to the biggest of them all: alchemy. For hundreds of years, many very smart people thought they were on the verge

I always like to compare holy grail problems to the biggest of them all: alchemy. For hundreds of years, many very smart people thought they were on the verge of becoming richer than kings but came up with bupkus until it was discovered that the means by which lead can be turned into gold involves nuclear reactions. It has been done. http://en.wikipedia.org/wiki/Nuclear_transmutation http://en.wikipedia.org/wiki/Nuclear_transmutation But what matters is the field from which you are approaching a problem. Tabletop chemistry was not enough. You needed nuclear physics. Maybe logic is not enough for AI. Maybe you need psychology, or some other field. In any event, great ideas don't care about probability. Google was a great idea, came in 1998, and kicked ass. I've thought I've had billion-dollar ideas for eight years now. It's always something world-changing, but I've been wrong so many times that when I come up with a good idea I take it with a sense of humor.