I don't like how in many contemporary films and series everybody is so curt and low affect, don't greet properly, barely look at each other. It's just not how people actually interact with each other. It's a cheap way of generating nerve for psychologically naive people.

It's underrated how fundamentally unstrategic the far left is, often attacking people they could build bridges to and bringing their opponents closer to each other by lumping them with each other.

As far as I can tell, no other major political group behaves like this.

A bit related to this thread

x.com/StefanFSchuber…

Unfortunate, but many tasks are harder to automate than it might initially seem.

Reality is surprisingly detailed:

johnsalvatier.org/blog/2017/real…

If you break etiquette rules in a way that annoys people, you often won't get feedback.

It can lead people to like you less; but they often won't tell you, instead just, e.g. avoiding you.

This lack of feedback can lead people to underrate the benefits of following such rules.

Many cultures with traditional family values certainly have or have had high fertility, so I don't think that is quite right as a general claim

Yes, though thanks to other rational investors such irrationality likely hurts you less than it does re many other decisions.

On the stock market, choosing at random hurts you less.

Many people defer more than they realise - including people who are epistemologically committed to thinking independently.

The pattern of their views (e.g. of other people's competence) is better explained by deference (e.g. to credentials or their ingroup) than by inside models

The EA-Progress studies war is here, and it's a constructive dialogue!

Clara Collier and Jason Crawford discuss what motivates and differentiates the two communities

(We're hoping Marc Andreessen doesn't read this and polarize everyone again)
asteriskmag.com/issues/06/the-…

New preprint and blog post!

We explore the social disincentives of warning about unlikely risks.

Many people are reluctant to warn about large but unlikely risks because they could look bad if the risk doesn’t occur

1/5

New preprint (with Spencer Greenberg 🔍, Alexander Grishin, Lewis & Lucius Caviola)!

We test the ability of LLMs and a specialised AI system to predict correlations between personality item pairs, comparing it to human experts and lay people.

For more details and results, see below!

Good set of graphs from an important book.

Total factor productivity growth is maybe the most striking.

This is not well-informed journalism

Indeed. Too many pretend to be something they're not. An age-old sport.