3 ms·

Nice, you're back. > I can for example show that a surge in the population of rabbits on a small island 'predicts' a surge of the population of their main pred

by Sakes 11y ago

Nice, you're back. > I can for example show that a surge in the population of rabbits on a small island 'predicts' a surge of the population of their main predator. That's the island ecosystem predicting a population bubble and then acts on those predictions. Intelligent? So I am assuming you are claiming to be part of the ecosystem? > That's beside the point though, because your example didn't show 'capitalism' making a prediction, it showed the Fed making a prediction. Not besides the point, it attacks a necessary part of my argument. Anyways, you have made some good points. I'll give it some thought and get back to you.

nevinera 11y ago
>So I am assuming you are claiming to be part of the ecosystem? In general, I am part of many systems (eco- and otherwise), but no; that section of reply was an argument against the idea that being able to 'predict' the future was a reasonable way to define intelligence. It doesn't really change the argument if you stick me on the island though - I behave intelligently, and that affects the behavior of the ecosystem, but that doesn't cause the ecosystem be 'intelligent'. It's interesting to have a debate this slowly, I feel like it's unusually civil :-)
- Sakes 11y ago
  Ya, I've been enjoying it :) I was worried you hadn't based upon your previous comment. I'm sorry I haven't returned to this yet, but will soon I promise. I think we should try to find out where we have agreement and where we begin to diverge. That should make the remainder of this discussion tighter.
  - nevinera 11y ago
    I believe our divergence is somewhere under the definition of 'intelligence'. The word is very ill-defined, to the point that we are reduced to describing it functionally in many cases, as with the Turing test. >Human intelligence is the intellectual capacity of humans, which is characterized by perception, consciousness, self-awareness, and volition. That's a very normal definition of intelligence. AI is defined (among many other way, naturally) as "the study of intelligent agents"; 'agency' is a requirement for an artificial intelligence. Since we started with the context of 'paperclip maximizers', I've been talking about intelligence in that context - in order to be studied as an intelligence, a thing must have 'agency', the ability to intentionally act. A thing that doesn't have agency can still have behavior, and that behavior can still be studied, but the behavior is emergent, not intentional - the system does not have agency unless it is capable of having goals, and of making decisions to achieve or progress toward them. In particular, capitalism isn't 'trying to maximize capital' - that's an effect it's (supposedly) having, but not an intentional one. It's a pretty clear emergent effect - if many of the actors in a system are trying to maximize their personal capital, then the system turning out to maximize net capital should surprise nobody. It's very equivalent to calling a diffusion chamber a 'maximizer' of entropy - a diffusion chamber does maximize entropy, but by calling it a 'maximizer' (in the context of 'paperclip maximizer'), you would ascribe agency to the chamber - it's not trying to maximize entropy, that's just something that it happens to do.

by Sakes 11y ago

nevinera 11y ago

>So I am assuming you are claiming to be part of the ecosystem? In general, I am part of many systems (eco- and otherwise), but no; that section of reply was an argument against the idea that being able to 'predict' the future was a reasonable way to define intelligence. It doesn't really change the argument if you stick me on the island though - I behave intelligently, and that affects the behavior of the ecosystem, but that doesn't cause the ecosystem be 'intelligent'. It's interesting to have a debate this slowly, I feel like it's unusually civil :-)

Sakes 11y ago
Ya, I've been enjoying it :) I was worried you hadn't based upon your previous comment. I'm sorry I haven't returned to this yet, but will soon I promise. I think we should try to find out where we have agreement and where we begin to diverge. That should make the remainder of this discussion tighter.
- nevinera 11y ago
  I believe our divergence is somewhere under the definition of 'intelligence'. The word is very ill-defined, to the point that we are reduced to describing it functionally in many cases, as with the Turing test. >Human intelligence is the intellectual capacity of humans, which is characterized by perception, consciousness, self-awareness, and volition. That's a very normal definition of intelligence. AI is defined (among many other way, naturally) as "the study of intelligent agents"; 'agency' is a requirement for an artificial intelligence. Since we started with the context of 'paperclip maximizers', I've been talking about intelligence in that context - in order to be studied as an intelligence, a thing must have 'agency', the ability to intentionally act. A thing that doesn't have agency can still have behavior, and that behavior can still be studied, but the behavior is emergent, not intentional - the system does not have agency unless it is capable of having goals, and of making decisions to achieve or progress toward them. In particular, capitalism isn't 'trying to maximize capital' - that's an effect it's (supposedly) having, but not an intentional one. It's a pretty clear emergent effect - if many of the actors in a system are trying to maximize their personal capital, then the system turning out to maximize net capital should surprise nobody. It's very equivalent to calling a diffusion chamber a 'maximizer' of entropy - a diffusion chamber does maximize entropy, but by calling it a 'maximizer' (in the context of 'paperclip maximizer'), you would ascribe agency to the chamber - it's not trying to maximize entropy, that's just something that it happens to do.