4 ms·
It may be a tough problem for the current state of AI, but one day I want to see to life a consequence engine for real-life decisions. For example, when a new l
by dzink 12y ago
It may be a tough problem for the current state of AI, but one day I want to see to life a consequence engine for real-life decisions. For example, when a new law is proposed, the engine interprets every stipulation, relates it to economic principles, legal decisions, historic analogies, etc, and produces probabilities of second, third and fourth orders of consequences that might result from the law as structured. There are many bad ideas and decisions made in government and elsewhere that could be avoided if something objectively evaluates the consequences of said politically or pork-motivated moves.
- walterbell 12y agoThis would be more tractable if there were multiple consequence graphs that could be overlaid on wikipedia. Prior to modelling possible futures, a good exercise is to model alternate pasts. Pick key turning points in history and identify variables where a small change in input (location, time, magnitude) would have resulted in a large change in output.