John Wentworth: That’s something I’ve been working on a fair bit lately is trying to figure out what exactly is needed. John Wentworth: Things that could exist, the big bottleneck right now is just getting more independent researchers to the point where they can do useful independent research in alignment. John Wentworth: If you want to tackle the core problems then independent research is clearly the way to go right now. Redwood is doing mildly interesting empirical things, but not even attempting to tackle core problems. That’s an important skill, we’re going to have to install that one, when you’re doing some math, you want to have a picture in your head of the prototypical example of the thing that the math is saying." That’s a very crucial load bearing skill. John Wentworth: So for instance, I was working with my apprentice recently just doing some math, and at one point I paused and asked him, "All right, sketch for me the picture in your head that corresponds to that math, you just wrote down." And he was like, "Wait, picture in my head? I wrote a post recently arguing that a big reason to study physics is that physicists in particular seem to have a bunch of illegible epistemic skills, which somehow get passed on to new physicists, but nobody really seems to make them legible along the way.

J.A. Robinson is perhaps the staunchest proponent of the position that the two are nonequivalent, arguing that there is a nonequivalence in meaning and that they fail to capture the same extension. There are two values assigned towards the letter C. Before the letters A, O, or U, it's pronounced like the C in CAT, as in CASA or CANTAR. This is the very same content that leads to the two definitions. ’s lots of other people who have followed MIRI to about the same extent that I have and have not converged to similar views to nearly the same extent, right? So you’re an independent researcher, right? I imagine this might be a deficit in the independent research landscape. Daniel Filan: Okay. So next I want to talk a little bit about how you interact with the research landscape. Little ones can find themselves intrigued in coloring and dress-up video games, while young children appreciate bicycle and automobile driving video games and a variety of arcades, when older people will just adore all the billiards, cards, puzzles and logic games and not to forget all the classic remakes like Super Mario, Pac Gentleman and so on.

John Wentworth: Daniel can see this from my face, but yeah. John Wentworth: So that’s obviously hugely valuable infrastructure for me personally. John Wentworth: Yeah. I assume that’s because they’re all defective somehow and do not understand things at all. John Wentworth: So the right now de facto answer is that’s most of the substantive research in AI alignment fits best with independent research. This includes side spinning, the right and left English, and others. Right now, we don’t even know what are the key questions to ask about alignment. And when I talk about tackling the core problems, I’m mostly talking about research directions which will plausibly get us to the point where we know which questions we need to ask or which things we even need to do in order to robustly build an AI which does what we want. So here are listed some facilities your health club must provide in order to be able to provide a comprehensive health and entertainment package. John Wentworth: We’re at the stage where we don’t know which questions we need to ask in order to do that. John Wentworth: So then did a few exercises on that and within a week or two, there was a very noticeable improvement as a result of that.

There are different kinds of sports like cycling, diving, hunting, billiards, fishing, golf, archery, shooting and a lot more. There’s a lot of people who’d like to do this, but they don’t really know what to do or where to start. So there’s a lot of currently not very legible skills that go into doing this sort of pre-paradigmatic research. Daniel Filan: Okay. So I expect that a lot of listeners to the show will be pretty familiar with AI, maybe less familiar with biology or economics. John Wentworth: If I’m thinking more about what would be useful for me personally, there’s definitely a lot of space to have people focused on distillation. John Wentworth: There’s no one that immediately jumps out as the person in economics or in biology who is clearly asking the analogous questions to what we’re doing here. John Wentworth: Yeah. So the obvious big ones currently are LessWrong / the Alignment Forum, and the Lightcone offices. Daniel Filan: Which are offices in Berkeley where some people work.

If you loved this write-up and you would like to receive much more facts concerning What is Billiards kindly go to our web site.