I hope you realize how hopeless this sounds. Historically speaking, human beings are exceptionally bad at planning in advance to contain the negative effects of new technologies. Our ability to control the adverse side-effects of energy production, for example, have been remarkably poor; decentralized market-based economies quite bad at mitigating the negative effects of aggregated short-term economic decisions. This should be quite sobering: the negative consequences of energy production are very slow. At this point we have had decades to respond to the looming crises, but a combination of ignorance, self-interest, and sheer incompetence prevents us from taking action. The unleashing of AI will likely happen in a heartbeat by comparison. It seems utterly naive to think that we can prevent, control, or even guide it.
Re: “The AI Problem, with Solutions”
I hope you realize how hopeless this sounds. Historically speaking, human beings are exceptionally bad at planning in advance to contain the negative effects of new technologies. Our ability to control the adverse side-effects of energy production, for example, have been remarkably poor; decentralized market-based economies quite bad at mitigating the negative effects of aggregated short-term economic decisions. This should be quite sobering: the negative consequences of energy production are very slow. At this point we have had decades to respond to the looming crises, but a combination of ignorance, self-interest, and sheer incompetence prevents us from taking action. The unleashing of AI will likely happen in a heartbeat by comparison. It seems utterly naive to think that we can prevent, control, or even guide it.