Reminded me of this: https://youtu.be/AIU9Q-9OzA0
Bonus points if you can get everyone to break out into acapella
Rachel Shu
Karma: 636
I don’t think so. Most EAs are not rats (and this is an extremely good thing!!!) but this means 80k is not the place to filter for LWer types when that is what you’re looking for
I’d point at myself but I’ve only complained about the problem :p https://rachelshu.com/2024/03/08/oceans-five.html
Deb Tannen is specifically who I had in mind! She’s most famous for her book on male vs female communication but if you read her other works such as on parent-child and friend-friend communication styles you get a good sense of the breadth of her framework.
Spencer Greenberg (who’s on here) also has a pretty substantial body of work at https://www.clearerthinking.org/
Type of guy who has not encountered sufficient girl autism
While broadly agreeing with David about the overall framing of the problem, it is interesting to see the metacognitive approach you give here reflected in the latest Claude Constitution (which has a lot of “we hope you will do what you think we really meant”), and I think does some credit to your insight.
There are a set of Moorean statements such as “there are no bugs crawling on me, but I believe there are”, which translate readily into comprehensible mental states. A rationalist might stress “but I alieve there are”.
There’s a whole category of intensifiers which implicate the real: genuinely, actually, really, truly, seriously, substantially, very, definitely
Then there are intensifiers which implicate the imaginary: fabulously, unbelievably, incredibly, fantastically, impossibly, miraculously
It’s genuinely incredibly interesting to me how compatible these usages are!
By the time I went to CFAR in 2019 this felt like it had already become the dominant flavor of inner-circle rationalist thinking, but then that inner circle kind of petered out in influence. The person I see carrying that torch most loudly in my current social atmosphere is Chris Lakin.
But overall rationality has been kind of quiescent imo! Ray posts good stuff, Duncan has his own thing, but it feels like we went from mid-2010s “rationalists talk a big game but don’t get anything done” to the mid-2020s most influential rationalists being too object-level busy to blog much about this metacognitive stuff.
My personal experience of life is that 70% of jam jars opened by myself and the women around me pass unremarkably without comment, and then another 10% even the guys can’t open without tools. Would be curious to get some empirical measurements on the actual variation in resistance between jars, even of the same brand. Could be extremely spiky!
But also I’d gladly pay an extra buck for jam to avoid needing to strategize how to open the damn thing 30% of the time!
Cease, Linch, to chide the follies of mankind,
Whose erring will, though seeming free, was wrought
In that first Forge whence reason, weak and warped,
Issued half-finished. Blame not Adam’s sons —
Arraign the Hand that shaped them so!
Yeah, I think the generous way to interpret the objector in this dialogue is as saying “I think you are underestimating how hard this problem is to fix”.
It’s not just that B is responsible instead of A, but that B has different constraints than A. Maybe you don’t see any constraints on A that would make the problem hard to solve. Yes—that is because the problem’s constraints are on B instead!
I think “the fact that you’re even complaining about X generically rather than explicitly X modulo B’s constraints this means you’ve thought about it less than I have” is an unfortunately common condescension.
Mox is the largest AI Safety community space in San Francisco. We’re fundraising!
Maybe this helps evade a set of jailbreaks where the user could otherwise create a situation where they convince Opus to break its safety measures in order to help Anthropic not lose money?
I actually think that it is both predictable and also a merit to our society that doctors are dumber than they used to be. Following Whitehead’s precept that “Civilization advances by extending the number of important operations which we can perform without thinking of them”, we’d ideally at once broaden the pool of potentially qualified medical practitioners and reduce the difficulty of succeeding at their profession.
If, in the future, any Homer Simpson can safely suture your wounds and chemotherapize your cancers, would you insist they had the reading skills of a medical student of the 1860s?
I’ll grant that the quotations from the centennial Etiquette are simply horrendous, but apart from its prose defects it should be noted that we currently live in a society in which the concept of ‘good breeding’ has no proper equivalent (big loss imo), social skill is defined more by charm than class (mixed bag), and business etiquette has been elevated to a minimal cross-cultural norm of tolerance and non-offensiveness (big win). So we should in general expect a modern treatment of manners to be vaguer, and more people to read other sorts of books for social advice.
The opposite is also frequently true, perhaps even for the very same people: being in a position of authority is also a reliable fix for akrasia. Common examples: “I have a kid now, my family relies on me”, “my community is depending on me”, “my employees need me to make the best decision for the company”.
Even in the example given, almost every officer in the military has both superiors and direct reports.
There are two types of bad scientific theories:
not wrong, but unnecessarily complicated for use in describing the phenomena it wants to describe, like the geocentric model.
actually wrong, like the emission theory of vision (the idea that we see by shooting rays from our eyes).
I don’t think it’s that useful to call them both false. Geocentrism is outdated in the same sense that woodblock printing is outdated: highly inefficient at the task it’s meant to do relative to better tools.
Ooh, one way that you could have it is that the human is actually solving problems by programming a bot to solve them, a bit like in Shenzhen I/O, and the bot starts to meet lookalike bots, that act according to the code you’ve written, but with an opposing goal? And they’re on the other side of the mirror so you can only change your own bot’s behavior
Why isn’t “speak your conscience (so people know you’re an honest person) but vote your constituency (to fulfill the role you are elected to)” a generally viable strategy?