3 ms·
I think you're conflating two different things: that such systems should generally be oriented towards maximizing human wellbeing (which I wholeheartedly suppor
by sp527 8y ago
I think you're conflating two different things: that such systems should generally be oriented towards maximizing human wellbeing (which I wholeheartedly support) and such systems should be vulnerable to human input/error (problematic in almost every conceivable situation).
Of course you can't swap every system we have for a fair and automated equivalent. You also run into problems around the question of what we should optimize for, which will be swayed by a person's political opinion. Ideal systems would probably lie at a balancing point between socialism and capitalism (as a model, equivalent to tuning parameters), but establishing that point and getting everyone to agree on it would be extremely challenging.
- panic 8y agoYeah, I think the question of what we should optimize for is the fundamental problem here. There's really nothing called "human wellbeing" that can be maximized. Everyone has different perspectives, values, goals, ideas, dreams, and so on. It's impossible to turn all these ways of seeing the world into a working system without involving the people themselves, especially given how people's perspectives change over time.