At what point do the economics and the scaling laws make AI no longer economically feasible? Like "in order to justify our capitalization we need to consume all the power in the universe territory"?
This needs fleshing out. Does the AI affect (motion-plan for, listen to, sense-from) new real-world objects, or is it just for everybody to compose all their communications with. If new objects, or new interactions with current objects, where does their energy come from? Increased economic activity is more work performed, which comes from more consumption and/or greater efficiency, and efficiency gains only get you so far.
At what point do the economics and the scaling laws make AI no longer economically feasible? Like "in order to justify our capitalization we need to consume all the power in the universe territory"?
This needs fleshing out. Does the AI affect (motion-plan for, listen to, sense-from) new real-world objects, or is it just for everybody to compose all their communications with. If new objects, or new interactions with current objects, where does their energy come from? Increased economic activity is more work performed, which comes from more consumption and/or greater efficiency, and efficiency gains only get you so far.
I am smart CEO of shipping company.
For our model to be successful we need 3km long ships, but unfortunately no one can build such ships. That's how smart and innovative we are.