Live data from Hacker News

Guess I'm a rationalist now

scottaaronson.blog

281–290 of 1001 posts

Re: Guess I'm a rationalist now

#281

Logic is an awesome tool that took us from Greek philosophers to the gates on our computers. The challenge with pure rationalism is checking the first principles that the thinking comes from. Logic can lead you astray if the principles are wrong, or you miss the complexity along the way. On the missing first principles, look at Aristotle. One of the history's greatest logicians came to many false conclusions. On miss…

Logic is the study of what is true, and also what is provable.

In the most ideal circumstances, these are the same. Logic has been decomposed into model theory (the study of what is true) and proof theory (the study of what is provable). So much of modern day rationalism is unmoored proof theory. Many of them would do well to read Kant's "The Critique of Pure Reason."

Unfortunately, in the very complex systems we often deal with, what is true may not be provable and many things which are provable may not be true. This is why it's equally as important to hone your skills of discernment, and practice reckoning as well as reasoning. I think of it as hearing "a ring of truth," but this is obviously unfalsifiable and I must remain skeptical against myself when I believe I hear this. It should be a guide toward deeper investigation, not the final destination.

Many people are led astray by thinking. It is seductive. It should be more commonly said that thinking is but a conscious stumbling block on the way to unconscious perfection.

Re: Guess I'm a rationalist now

#282

The article made me think deeper about what rubs me the wrong way about the whole movement I think there is some inherent tension btwn being "rational" about things and trying to reason about things from first principle.. And the general absolutist tone of the community. The people involved all seem very... Full of themselves ? They don't really ever show a sense of "hey, I've got a thought, maybe I haven't considere…

They remind me of the "Effective Altruism" crowd who get completely wound up in these hypothetical logical thought exercises and end up coming to insane conclusions that they feel trapped in because they got there using pure logic. Not realizing that their initial conditions were highly artificial so any conclusion they reach is only of academic value. There is a term for this. "Getting stuck up your own butt." It wo…

> They remind me of the "Effective Altruism" crowd

Isn't there a lot of overlap between the two groups?

I recently read a great book that examines these various groups and their commonality: More Everything Forever: AI Overlords, Space Empires, and Silicon Valley's Crusade to Control the Fate of Humanity by Adam Becker. Highly recommended.

Re: Guess I'm a rationalist now

#283

Earlier quoted context omitted.

> They don't really ever show a sense of "hey, I've got a thought, maybe I haven't considered all angles to it, maybe I'm wrong - but here it is". Aren't these the people who started the trend of writing things like " epistemic status: mostly speculation " on their blog posts? And writing essays about the dangers of overconfidence? And measuring how often their predictions turn out wrong? And maintaining webpages tit…

Weirdly enough, both can be true. I was tangentially involved in EA in the early days, and have some friends who were more involved. Lots of interesting, really cool stuff going on, but there was always latent insecurity paired with overconfidence and elitism as is typical in young nerd circles. When big money got involved, the tone shifted a lot. One phrase that really stuck with me is "exceptional talent". Everyone…

I don't think this validates the criticism that "they don't really ever show a sense of[...] maybe I'm wrong".

I think that sentence would be a fair description of certain individuals in the EA community, especially SBF, but that is not the same thing as saying that rationalists don't ever express epistemic uncertainty, when on average they spend more words on that than just about any other group I can think of.

Re: Guess I'm a rationalist now

#284
post #199

Earlier quoted context omitted.

37% of children obviously do not experience a CPS investigation before age 18.

not what i am speaking to. I don't know the number, and neither do you. you'd have to call those 5000 CPS caseworkers and ask them what their caseload is (it's 69 per caseworker on average across the US. that's a third of a million cases, in aggregate across all caseworkers) my wife's caseload (adults) "floats around fifty."

> not what i am speaking to

My misunderstanding then - what are you speaking to? Even reading this comment, I still don't understand.

Re: Guess I'm a rationalist now

#285

The article made me think deeper about what rubs me the wrong way about the whole movement I think there is some inherent tension btwn being "rational" about things and trying to reason about things from first principle.. And the general absolutist tone of the community. The people involved all seem very... Full of themselves ? They don't really ever show a sense of "hey, I've got a thought, maybe I haven't considere…

I've always seen the breathless Singularitarian worrying about AI Alignment as a smokescreen to distract people from thinking clearly about the more pedestrian hazards of AI that isn't self-improving or superhuman, from algorithmic bias, to policy-washing, to energy costs and acceleration of wealth concentration. It also leads to so-called longtermism - discounting the benefits of solving current real problems and fo…

>s a smokescreen to distract people from thinking clearly about the more pedestrian hazards of AI that isn't self-improving or superhuman,

Anything that can't be self-improving or superhuman almost certainly isn't worthy of the moniker "AI". A true AI will be born into a world that has already unlocked the principles of intelligence. Humans in that world would be capable themselves of improving AI (slowly), but the AI itself will (presumably) run on silicon and be a quick thinker. It will be able to self-improve, rapidly at first, and then more rapidly as its increased intelligence allows for even quicker rates of improvement. And if not superhuman initially, it would soon become so.

We don't even have anything resembling real AI at the moment. Generative models are probably some blind alley.

Re: Guess I'm a rationalist now

#286
post #263
post #218

Earlier quoted context omitted.

And today the railroad system in the USA sucks compared to other developed countries and even China. It turns out that boom-and-bust capitalism isn’t great for building something that needs to evolve over centuries. Perhaps American AI efforts will one day be viewed similarly. “Yeah, they had an early rush, lots of innovation, high valuations, and robber barons competing. Today it’s just stale old infra despite the h…

> And today the railroad system in the USA sucks compared to other developed countries and even China. Nonsense. The US has the largest freight rail system in the world, and is considered to have the most efficient rail system in the world to go along with it. There isn't much in the way of passenger service, granted, but that's because people in the US aren't, well, poor. They can afford better transportation option…

> There isn't much in the way of passenger service, granted, but that's because people in the US aren't, well, poor. They can afford better transportation options.

This is so misguided view... Trains (when done right) aren't "for the poor", they are great transportation option, that beats both airplanes and cars. In Poland, which isn't even close to the best, you can travel between big cities with speeds above 200km/h, and you can use regional rail for your daily commute, both those options being very comfortable and convenient, much more convenient than traveling by car.

Re: Guess I'm a rationalist now

#287

The article made me think deeper about what rubs me the wrong way about the whole movement I think there is some inherent tension btwn being "rational" about things and trying to reason about things from first principle.. And the general absolutist tone of the community. The people involved all seem very... Full of themselves ? They don't really ever show a sense of "hey, I've got a thought, maybe I haven't considere…

They remind me of the "Effective Altruism" crowd who get completely wound up in these hypothetical logical thought exercises and end up coming to insane conclusions that they feel trapped in because they got there using pure logic. Not realizing that their initial conditions were highly artificial so any conclusion they reach is only of academic value. There is a term for this. "Getting stuck up your own butt." It wo…

People who confuse the map for the territory.

Re: Guess I'm a rationalist now

#288
post #218
post #183

Earlier quoted context omitted.

When railroads were first being built across the continental USA, those companies also had high valuations (for the time). Most of them ultimately went bankrupt or were purchased for a fraction of their peak valuation. But the tracks remained, and many of those routes are still in use today.

And today the railroad system in the USA sucks compared to other developed countries and even China. It turns out that boom-and-bust capitalism isn’t great for building something that needs to evolve over centuries. Perhaps American AI efforts will one day be viewed similarly. “Yeah, they had an early rush, lots of innovation, high valuations, and robber barons competing. Today it’s just stale old infra despite the h…

I think it's unlikely that AI efforts will go as railroads have. I think being an AI foundation model company is more like being an airplane builder than like a railway company, since you develop your technology.

Re: Guess I'm a rationalist now

#289
post #108

Earlier quoted context omitted.

"Reductionist" is usually used as an insult. Many people engaged in intellectual pursuits believe that reductionism is not a useful approach to studying various topics. You may argue otherwise, but then you are on a slippery slope towards politics and culture wars.

I would not be so sure. There are many fields where reductionism was applied in practice and it yielded useful results, thanks to computers. Examples that come to mind: statistical modelling (reduction to nonparametric models), protein folding (reduction to quantum chemistry), climate/weather prediction (reduction to fluid physics), human language translation (reduction to neural networks). Reductionism is not that u…

I meant that the word "reductionist" is usually an accusation of ignorance. It's not something people doing reductionist work actually use.

Re: Guess I'm a rationalist now

#290

Just to confirm, this is about: https://en.wikipedia.org/wiki/Rationalist_community and not: https://en.wikipedia.org/wiki/Rationalism right?

Along these lines I am sort of skimming articles/blogs/websites about Lightcone, LessWrong, etc, and I am still struggling with the question...what do they DO?
Post reply on HN