LessWrong 2.0 Reader

Ah, yes - bargaining solutions that ignore or hide a significant underlying power disparity are rampant in wishful-thinking academic circles, and irrelevant in real life.  That's the context I was missing; my confusion is resolved.  Thanks!

This is basically because of the value of the long tail.

Automating 90% or 50% of your job is not enough to bring in lots of the value proposition of AI, because then the human becomes a bottleneck, which becomes especially severe in cases requiring high speed or lots of context.

@johnswentworth [LW · GW] talks about the issue here:

https://www.lesswrong.com/posts/Nbcs5Fe2cxQuzje4K/value-of-the-long-tail [LW · GW]

Yeah I said that to Matt Barnett 4 months ago here [EA(p) · GW(p)]. For example, one man's "avoiding conflict by reaching negotiated settlement" may be another man's "acceding to extortion". Evidently I did not convince him. Shrug.

(only if you pull the papers every day!)

Not intended to be an insult: I find it astonishing that it's possible to not notice that people mostly eat for enjoyment, and that eating to sustain oneself is not the norm. (Of course, this is coming from a person who eats mostly for enjoyment, and couldn't imagine that some people mostly eat for enjoyment.)

In particular, why do you think that so many people were overweight/obese? You flagged this with your remark about the fact that people mostly "cannot just-", but still, mechanistically, what did you think was causing most people to overeat?

Because that's what investors want. From observations at my workplace at a B2C software company^[1] and from what I hear from others in the space, there is tremendous pressure from investors to incorporate AI and particularly "AI agents" in whatever way is possible, whether or not it makes sense given the context. Investors are enthusiastic about "a cheap drop-in replacement for a human worker" in a way that they are not for "tools which make employees better at some tasks"

The CEOs are reading the script they need to read to make their boards happy,. That script talks about faster horses, so by golly their companies have the fastest horses to ever horse.

Meanwhile you have tools like Copilot and Cursor which allow workers to vastly amplify their work but not fully offload it, and you have structured outputs from LLMs allowing for conversion of unstructured to structured data at incredible scales. But talking about your adoption of those tools will not get you funding, and so you don't hear as much about that style of tooling.

1. ^{^}

   Obligatory "Views expressed are my own and do not necessarily reflect those of my employer"

In this case prediction markets will be predictably over-optimistic, and expert consensus is very split.

A computable hypervisor can't run an uncountable number of different VMs. You seem to be talking about a countable number of segments of the real line, with identical VMs in each segment. That gives you a countable number of different. VMs, but it's not really a bijection, because it's not general. And it's only an abstraction that a segment consists of uncountable identical VMs, not just one with a rational measure

No matter how much I try, I just cannot force myself to buy the premise of replacement of human labor as a reasonable goal. Consider the apocryphal quote: 

If I had asked people what they wanted, they would have said faster horses. –Henry Ford

I'm clearly in the wrong here, because every CEO who talks about the subject talks about faster horses^[1], and here we have Mechanize whose goal is to build faster horses, and here is the AI community concerned about the severe societal impacts of digital horse shit.

Why, exactly, are all these people who are into accelerating technical development and automation of the economy working so hard at cramming the AI into the shape of a horse?

1. ^{^}

   For clarity, faster horses here is a metaphor for the AI just replacing human workers at their existing jobs.

It sometimes takes me a long time to go from "A is true", "B is true", "A and B implies C is true" to "C is true".

I think this is a common issue with humans, for example I can see a word such as "aqueduct", and also know that "aqua" means water in Latin, yet fail to notice that "aqueduct" comes from "aqua". This is because when I see a word it does not trigger a dynamic that searches for a root.

Another case is when the rule looks a bit different, say "a and b implies c" rather than "A and B implies C" and some effort is needed to notice that it still applies.

I think an even more common reason is that the facts are never brought in working memory at the same time, and so inference never happens.

All this hints to a practical epistemological-fu: we can increase our knowledge simply by actively reviewing our facts, say every morning, and trying to infer new facts from them! This might even create a virtuous circle, as the more facts one infers, the more facts one can combine to generate more inferences.

On the other hand there is a limit to the number of facts one can review in a given amount of time, so perhaps a healthy epistemological habit to have is to trigger one's inference engine every time one learns a new (significant?) fact.