Stuff I Found Interesting In August
1) This Raj Chetty study on Social Capital:
The share of high-SES friends among individuals with low SES—which we term economic connectedness—is among the strongest predictors of upward income mobility identified to date.
I haven’t figured out how likely this is to be causal yet, please don’t assume that this is definitely mostly a causal effect. Here is a useful NYT piece on the study.
2) Understanding Jane Street.
3) If Holden Karnofsky is correct that we are likely in the most important century, does that make the claim that we are all in a simulation more likely?
4) Scott Alexander on The Line:
Is this just some crazy attempt to build hype, like when Elon Musk says the next Tesla definitely will have full-self-driving ability? I don’t think so. Saudi Crown Prince Mohammed bin Salman is obsessed with Neom and very vain; I don’t think he would deliberately promise impossible things knowing that he will be embarrassed later when they don’t work out.
5) This piece in The Times [paywalled]: ‘If you accept the standard view of education in Britain, two black boys have the same chance of success. They don’t’, and a thread by Akala responding:
6) Friend of the blog Peter McLaughlin reviews Rationalism, Pluralism, and Freedom.
7) Caplan does stand-up comedy:
It’s bad (sorry Bryan), but it’s not really terrible! I’ve been to a ton of open mic comedy and believe me when I say that ‘not really terrible’ is much better than at least 30% of people doing open mic comedy.
NOTE: WWOTF Coverage and other EA stuff begins here, skip forward if you’ve read it all already.
8) MacAskill in the NYT (yes, I know you’ve seen it already, I just don’t care):
Suppose that I drop a glass bottle while hiking. If I don’t clean it up, a child might cut herself on the shards. Does it matter when the child will cut herself — a week, or a decade, or a century from now? No. Harm is harm, whenever it occurs.
9) And then a profile of MacAskill (plus a lot more) in the New Yorker. This one is really, really worth reading.
10) And another one in Time. Read at least one of these three - the New Yorker one is the most worthwhile IMO. This one is great too. NYT piece if you haven’t read and don’t intend to read the book. MacAskill on Tyler’s podcast and Ezra’s podcast are also worthwhile listens. There, I’m done!
11) FDB criticises EA.
12) Friend of the blog Eli Lifland on the relationship between caring about future people and believing that x-risk reduction should be our top priority. His blog in general is worth reading.
WWOTF Coverage and other EA stuff ends here.
14) Nice profile of the boss of Starling bank, in lunch with the FT [paywalled].
15)
16) Robert Long on why you might want to care about AI sentience.
17) Apparently the energy price signal is more important than you would think at this point, if this article from The Economist is to be believed.
18) I’ve been playing with Midjourney (the poor man’s DALL-E) a lot, and I think it will serve as the art for most of my blog posts from now one - including this one.
19) Maxim Lott on the Covid Fudge factor.
20) Julian Hazell has started a Substack, which you should probably subscribe to.
21) Interesting piece in the FT [paywalled] about voice actors being replaced by AI. I expect this to keep happening to more professions (who doesn’t?). I wonder how long I can keep writing this blog until AI is clearly much better than me.
22) This review of road safety as a cause area was interesting. I was interested in writing up a long piece on road safety for the OpenPhil cause exploration prize, but looks like someone else got there first! I chucked a few quid to AIP via GlobalGiving, who wrote that post and seem great.
There is nothing at all in the Chetty stuff that even implies causation, let alone gives evidence for it. It mostly tracks racial segregation, which obviously may itself cause serious problems, but Chetty gives us no extra reason to judge that cross group FRIENDSHIPS are the reason that segregation is bad (or indeed that segregation itself is the problem).
Also, for what it's worth, against the Economist piece here's Duncan Weldon arguing that price signals have done about all they can do and rationing is basically inevitable if the government wants to avoid a truly massive recession: https://duncanweldon.substack.com/p/in-the-bleak-midwinter?utm_source=substack&utm_medium=email
I think the crux of the disagreement here is in the degree to which household economising can be depended on to insulate individuals and the economy from the energy shock. The Economist piece explicit argues that 'households and industry will adapt more to higher prices', but I incline to the position that really there's nothing households can do to respond to these insane signals - the degree to which people would need to cut back to rationally economise would kill them. Cf. https://twitter.com/thhamilton/status/1563886341275926528?t=wlPYpgR_lC3pNIgoLs93wA&s=19