Douglas Hofstadter

And I think part of what these things do when I read them, and hopefully as we discuss them and l- live with these, it's the idea that this gives you some solace, that you don't beat yourself up too much because there are some things, the laws of whatever you wanna call it, physical truths that impact us, and if we fight them, we'll be a lot less happy and probably a lo- a, a, a little less successful.

So, um, the next, uh, law I would like to discuss is Hofstadter's law, named after Douglas Hofstadter, the son of a noble laureate physicist.

So he obviously had big shoes to fill.

You sound like you're so full of shit when you say these facts.

Okay.

God, I wish I was that creative.

Um, so Hofstadter's law is it always takes longer than you expect, even when you account for Hofstadter's law.

So that, this one's brilliant because it's basically saying things will take longer than you expect them to, even including the fact that you're gonna think they're gonna be longer.

Note, an idea can be valuable even if it isn't scientific.

Hofstadter's law.

It always takes longer than you expect, even when you take into account Hofstadter's law.

Now, Hofstadter framed his law in terms of time, but I found it also applies to money.

When we were doing some renovations on our apartment, we were advised to add an extra 20% to our budget as a cushion for hidden problems.

But then it actually cost 20% on top of that.

And sometimes by philosophers, yeah, like Plato or Plotinus is a similar abstract essence that lives on after death and sometimes precedes death.

And then sometimes it's used to describe a quality of consciousness that That is a unique pattern, and that's the way that Douglas Hofstadter uses it.

Gutel, Escher, Bach, soul is an emergent property.

It might disintegrate at your death, but it isn't reducible to any single biological dimension of you.

But if all math is is proof, and if all a proof does is build upon indisputable axioms in a purely logical way, and if every proof is a question with only two answers, yes or no, true or false, then how can any question be ambiguous or difficult to understand? All you need to do is remember the correct axioms or proofs and apply them in the right way at the right time.

The cognitive and computer scientist Douglas Hofstadter, best known for his concept of strange loops as explored in his tome Girdle Escher Bach, An Eternal Golden Braid, gave one answer to this question in a 2012 essay, which went lightly viral over the past year.

In response to a question from a reader of the journal Mathematics in School about the experience of being a distinguished mathematician, he first demurred, protesting his mathematical bona fides.

And then he went further, writing of the traumatic experience of waking up to his own mathematical inadequacy and limitation.

self? I don't have a precise answer, but I'll tell you what that brings up for me, which is, um, that when people ask me, ask me what the, what's the book that influenced my life more than any others.

And I know I'm supposed to say some business book or something, but I always say, go to Escher Bach, which is an extremely nerdy book by Douglas Hofstadter, um, who is a professor at University of Indiana and his book, which became like kind of within the like CS world and stuff.

It's reasonably widely read.

I mean, my understanding is the number of people that start the book versus finish, there's a big drop off, but it's 800 pages of like math and parables and stuff.

And this idea of, you know, whether it's through the screen, like the digital ether or sort of like the screen of the TV or the cinema.

that there is this kind of sort of like seance happening and there's a bit in the film where i think she's like scrolling past all the videos of people sort of like taking the challenge and the symptoms they're reporting and there's a couple of videos in the background for this thing called like what is the strange loop and uh when i was a teenager i was like very obsessed with this sort of like um this sort of like theorist of the mind philosophy of the mind called Douglas Hofstadter.

And he did this book called an eternal golden braid, which is about sort of like the idea of these sort of like loops.

And he fleshed out this idea of the strange loop theory, which is like, I guess the most common paradoxical image of it is like, it's the chicken and the egg, or it is sort of like MC Escher, the hands drawing the, each other right and i kind of like i really thought about that and my first memories where like you can't quite place like what what where which one starts and begins and ends and i think when you grow up watching films like so much of your formative memories end up being in this kind of like paradoxical tension with each other weirdly enough teenage sex and death was um, a very sort of like strange invocation of a lot of personal memories for me because my first boyfriend was actually a slasher scholar.

The, the... [laughs] It wasn't, it's not Sand, but he-

You're Hofstadter

... he's just a guy, he's a, he's another guy researching.

Yeah, he's a Hofstadter guy.

He was a guy researching AI.

But he told me it's an alien mind two, two years ago.

And the last question, what book changed your life?

Hands down for me, that book was Douglas Hofstadter's, Gertl Escher Bach, which I read while I was in high school.

And that was the kind of, the end of high school was like the very early days of the internet.

And so I was able to email Douglas Hofstadter before everyone on the planet was emailing Douglas Hofstadter because just this medium allowed me this way in and was able to get a response from him where, you know, what I was asking is, how do I study? Like, I want to study music and cognitive science.

How do I do this? Right.

And he gave me pointers and, you know, the rest unfolded from there.

On AI, let's not replicate the gain-of-function failures.

Scott Aronson argues that LLMs are a definitive counterexample to the strange loop and self-referentiality hypothesis of Douglas Hofstadter.

This seems correct.

Were the steps Anthropic took to pause some activities, as I discussed yesterday, comparable to what OpenAI did? Oliver Haberker argues it was not comparable at all.