Live data from Hacker News

Stop Fearing Artificial Intelligence

techcrunch.com

91–95 of 95 posts

Re: Stop Fearing Artificial Intelligence

#91
post #84
post #83

Earlier quoted context omitted.

> The Church–Turing thesis directly says that computers can do anything humans can, and it's pretty well accepted. Huh? It doesn't say anything close to that. That wouldn't even be mathematics. > Physical laws, as far as we can tell, are Turing computable. Not true either. Where are you getting this idea? I accept the materialistic thesis. I think people are conflating this thesis and the statement of whether human t…

I just looked over the Wikipedia on the Church Turing thesis and digital physics, and I still think that's right. The thesis says that a computer can compute anything a human can. You're right that it can't be proven, because "what a human can do" isn't formalized, but it still has almost universal acceptance. (The "strong" version does not have as much acceptance.) As for the universe being Turing computable, I thin…

I think I have to be harsh to get my point across: what you are saying is nonsense.

I will use this as one example, but you are equally confused on a number of issues, to the point where the conversation isn't productive. I thought you were asking honest questions to learn something, but you appear to have some preconceived notions which you will hold onto in face of contradictory evidence.

I said: "Computers, minds, and the universe are different things that work by different laws".

You said: "The Church–Turing thesis directly says that computers can do anything humans can"

It absolutely doesn't say that. Nor does the Wikpedia article say that it does.

Wikipedia says:

"In simple terms, the Church–Turing thesis states that a function on the natural numbers is computable in an informal sense (i.e., computable by a human being using a pencil-and-paper method, ignoring resource limitations) if and only if it is computable by a Turing machine."

Maybe you are confused by the mention of human beings in that sentence. That is irrelevant -- it would only be relevant if doing calculations on pencil and paper are the ONLY things that humans can do. Of course humans can do lots of other things, like make movies and do their taxes.

If you want to philosophise, that's OK. If you want to talk about mathematics, you can't "think it's right". Your opinion doesn't matter. You are confusing philosophy and mathematics.

And you are not doing Wikipedia justice: http://en.wikipedia.org/wiki/Church%E2%80%93Turing_thesis#Ph...

It's accurate in saying whether the universe can be simulated by a computer is an open question. That's what I'm saying. I'm not denying it. I'm saying you cannot assume it's true to make FURTHER philosophical arguments, and then expect me not to object!

You would be basing your argument on false premises. I guess it could be worse -- you could base entire 300 page books on false premises.

I think it's good to be honestly interested and excited about these subjects, but it's clear you have a lot to learn. Sorry to be harsh, but I would like to direct your interest to better sources. You seem to be learning from sources that doesn't understand the primary material themselves. Even Wikipedia is an OK source, and quite accurate in this case, but you are not reading it correctly.

Re: Stop Fearing Artificial Intelligence

#92
post #81

Earlier quoted context omitted.

> So at the very least, it can think in one day what a human can think in a million days. You are simply reiterating the original position that I was devils-advocating against. It is not proven that a million days of work equals superintelligence, even though obviously it would be more effective than 1 day. If some activity is exponential, like if it takes 10^x days to accomplish, then a human could solve problems of…

>It is not proven that a million days of work equals superintelligence, even though obviously it would be more effective than 1 day. >This is not hypothetical; the best known complete (not approximate) solutions to things like the traveling salesman problem, or to fitting odd sized containers into a container ship, are in fact exponential like that as far as we currently know. Why would something need to solve proble…

I am merely playing Devil's Advocate. I personally think that speedups of 1000-fold and million-fold etc. are a really big deal.

"Quantity has a quality all its own."

Nonetheless, it remains possible that there is a difference between fast intelligence solving things faster than we can, versus a superintelligence (from extraterrestrial aliens or something) that can create superior inventions or art or math or whatever that humans couldn't no matter how much time was allowed.

Time is not always the issue.

Anecdote from the Feynman biography "Genius":

A physicist once said that there were two kinds of genius. The first kind was the sort that a good but non-genius physicist could imagine that they, too, could have achieved the same results as the genius, if only they could work on just that for decades or even lifetimes.

But then he said that Feynman was the second kind of genius, whose results seemed to have been created by magic, and that ordinary physicists could not imagine achieving no matter how long they worked at it.

The key thing here is that some pretty smart people already believe that there are at least two kinds of very high intelligence, one of which is not related to speed.

So my point is merely that the same might apply to these other topics. We know so little about how to create a mind that there simply is no proof one way or the other.

And that latter is an objective fact. There is no existing algorithms for creating human-level minds that merely needs more cpu + ram + disk to work, there just isn't. So we don't know. Different people choose to believe one way or the other.

That's all.

Re: Stop Fearing Artificial Intelligence

#93
post #90

Earlier quoted context omitted.

Oh boy, good question. First, I'm not an AI expert, I don't even casually practice AI. I'm not a machine learning expert. I am a statistician that applies machine learning methods a handful of times per year. That to say --- I find Bayes rule to be eerily similar to the way I think a human mind learns. We operate with degrees of belief, we couple those beliefs with new information and update our beliefs. The word tax…

What I was getting at is that any no-free-lunch objection you make to learning algorithms in general also applies to the specific learning algorithm in our heads -- and yet we manage to function well enough to have this conversation.

Very good point. :-)

Re: Stop Fearing Artificial Intelligence

#94
post #35

I don't fear "AI". What I fear is AI being used as a replacement for human judgment or as a justification for greed and/or bigotry. We already have this problem, where decisions that should be made by the humans involved are restricted under the "policy" excuse of bureaucracy, blind to the reality of unusual situations that require an exception in the usual routine. I fear AI because it is harder to fight complicated…

They can't fool me, it's small perturbations in the 'Kill n to warn p,' policy and wrong packets in port -3 all the way down. The TC author seems to have a PhD in his field and never ever run into ...well, Agile book examples, of course...a case where modifying one's own rules turns a corner in either the comic, Malthusian or Hadoop renormalization cases. Now, I want computers to have new art up every time I turn around, but there's nothing to say making Jibbers Crabst come around is going to break homology.

Re: Stop Fearing Artificial Intelligence

#95

I fear AI far less than I fear humans. First, twenty thousand years or more of human history is about us trying to make other humans into machines: slavery, rape and forced prostitution, political coercion, war, overreaching prison systems and draconian punishments for minor crimes. We're not very nice to each other. Rather, we're controlling assholes as a species. Most of the ones who get power in organizations smal…

An idea that I got after reading Guns, Germs, and Steel was that at a basic level, killing other humans is the correct response to most threats, annoyances, or disagreements. Manager wants to micromanage with story points? Kill him. Problem solved.

Now, clearly, if everybody is killing everyone beyond their immediate family, it is hard to have any sort of larger society. Living and collaborating under constant threat of immediate death wouldn't work. So we have a large set of social adaptations to limit the killing and restrain ourselves.

My point here, is that 'why do human kill other humans' is a less interesting question than 'why are the social adaptations that prevent killing not taking affect?' The problem of fighting isn't a moral failing, or some flaw in the human soul. Killing makes sense within the decision making scope of our conscious brains, and is the result of inadequate build up of collective structures of inhibition that are required to have larger groups of people in closer contact with each other, and as the groups get bigger and the contact becomes closer, we need to engineer new schemes for inhibiting the 'lets just kill them off and solve this once and for all' urge.

Post reply on HN