Also one Spencer recorded with me: “Lines of Retreat and Incomplete Maps”. Not sure why it isn’t above; maybe was from earlier than the ones listed.
AnnaSalamon
Karma: 11,705
monitoring for symptoms and taking a large dose of antibiotics within ~48 hours of symptoms is extremely effective.
I’m confused about this. Can you say more about what your threshold is for “extremely effective,” or why you think so? Wikipedia states: “People who receive recommended antibiotic treatment within several days of appearance of an initial EM rash have the best prospects.[106] Recovery may not be total or immediate. The percentage of people achieving full recovery in the United States increases from about 64–71% at end of treatment for EM rash to about 84–90% after 30 months; higher percentages are reported in Europe.[171][172] Treatment failure, i.e. persistence of original or appearance of new signs of the disease, occurs only in a few people.[171] Remaining people are considered cured but continue to experience subjective symptoms, e.g. joint or muscle pains or fatigue.[173] These symptoms usually are mild and nondisabling.[173]”
This leaves me thinking that even with rapid antibiotics, the debilitation per (infection that causes a rash) is significant.
Yes, this is something of a crux for me—what are our odds of noticing, if we’re paranoid? Folks say a rash is only present in 70-80% of cases; do you have stats on how often the rest is present? (How would one generate those stats?)
Jefftk, can you say more about the people who know who didn’t initially get the symptoms but later got arthritic symptoms, and about how many people you know who got Lyme altogether, and how sure you are that the arthritic symptoms are from Lyme?
Thanks. I just bounced to LW after getting stuck in a tricky bit of writing, and found this helpful for where I was stuck.
I think the main things I found helpful from your post just now were:
the examples, which for me recalled the habit of righting a wrong question; and
the explicit suggestion that I could take the spirit of “righting a wrong question”, or “dissolving a question”—call it a “virtue”—and steer toward it in the way I might steer toward curiosity or other virtues.
I agree an algorithm could do as you describe.
I don’t think that’s what’s happening in me or other people. Or at least, I don’t think it’s a full description. One reason I don’t, is that after I’ve e.g. been camping for a long time, with a lot of room for quiet, it becomes easier than it has been to notice that I don’t have to see things the way I’ve been seeing them. My priors become “less stuck”, if you like. I don’t see why that would be, on your (zhukeepa’s) model.
Introspectively, I think it’s more like, that sometimes facing an unknown hypothesis (or rather, a hypothesis that’ll send the rest of my map into unknownness) is too scary to manage to see as a possibility at all.
One interesting datum about willpower (which I’ve observed repeatedly in many people and contexts; not sure if it’s officially documented anywhere) is that it’s much easier to take a fully scripted action than to take an action that requires creatively filling in details.
For example, suppose several people are out trying to do “rejection therapy” (a perhaps-dubious game in which folks make requests of strangers that are likely to be rejected, e.g. “Can I listen to the walkman you’re listening to for a moment?” or “Can we trade socks?”). Many many people who set out to ask a stranger such a question will… “decide” not to, once the stranger is actually near them. However, people who have written down exactly which words they plan to stay in what order to exactly which stranger, with no room for ambiguity, are (anecdotally but repeatedly) more likely to actually say the words. (Or to do many other difficult habits, I think.)
(I originally noticed this pattern in undergrad, when there was a study group I wanted to leave, but whose keeper I felt flinchy about disappointing by leaving. I planned to leave the group and then … didn’t. And the next week, planned again to leave the group and then didn’t. And the third week, came in with a fully written exact sentence, said my scripted sentence, and left.)
Further examples:
It’s often easier to do (the dishes, or exercise, or other ‘difficult’ tasks) if there’s a set time for it.
Creativity-requiring tasks are often harder to attempt than more structured/directions-following-y tasks (e.g. writing poetry, especially if you’re “actually trying” at it; or attempting alignment research in a more “what actually makes sense here?” way and a less “let me make deductions from this framework other people are using” way; or just writing a blog post vs critiquing one).
— I’ve previously taken the above observations as evidence for the “subparts of my mind with differing predictions” view — if there are different bits of my (mind/brain) that are involved in e.g. assembling the sentence, vs saying the sentence, then if I need to figure out what words to say I’ll need whatever’s involved in calling the “assemble sentences” bit to also be on board, which is to say that more of me will need to be on board.
I guess you could also try this from the “Bayesian evidence” standpoint. But I’m curious how you’d do it in detail. Like, would you say the prior against “moving muscles” extends also for “assembling sentences”?
The basic idea of a trapped prior is purely epistemic. It can happen (in theory) even in someone who doesn’t feel emotions at all. If you gather sufficient evidence that there are no polar bears near you, and your algorithm for combining prior with new experience is just a little off, then you can end up rejecting all apparent evidence of polar bears as fake, and trapping your anti-polar-bear prior. This happens without any emotional component.
I either don’t follow your more general / “purely epistemic” point, or disagree. If a person’s algorithm is doing correct Bayesian epistemology, a low prior of polar bears won’t obscure the accumulating likelihood ratios in favor of polar bears; a given observation will just be classified as “an update in favor of polar bears maybe being a thing, though they’re still very very unlikely even after this datapoint”; priors don’t mess with the direction of the update.
I guess you’re trying to describe some other situation than correct Bayesian updating, when you talk about a person/alien/AI’s algorithm being “just a little off”. But I can’t figure out what kind of “a little off” you are imagining, that would yield this.
My (very non-confident) guess at what is going on with self-reinforcing prejudices/phobias/etc. in humans, is that it involves actively insulating some part of the mind from the data (as you suggest), and that this is not the sort of phenomenon that would happen with an algorithm that didn’t go out of its way to do something like compartmentalization.
If there’s a mechanism you’re proposing that doesn’t require compartmentalization, might you clarify it?
I find this helpful; thank you. I wonder if this is part of what was going on in the overcaution-about-covid dynamic mingyuan describes.
So there’s definitely a measure of social inertia there that has nothing to do with fear of COVID.
My experience is that fear, or at least fear that is in the background and that I am dissociated from, creates social inertia and other inertia for me. (Also grief that I am dissociated from.)
AFAICT, fear (especially fear as a background that is just always there, for months and months) has huge effects on me and many others that are bad for thinking, initiative, activated caring, and real companionship (or being conscius at all, sort of), and that it requires actively training courage or bravery or action/initiative/activated-caring to overcome this. I notice this a lot in the AI risk context, and sometimes in the “what’s happening to America / the West?” context, and also most times that somebody is e.g. afraid they have cancer.
My personal experience of the covid year has been good; but your post seems to me to have a lot in it that bears on the broader thing about fear-in-general, and I think talking more about the detailed effects of fear in the context of this post, covid, AI risk or any other context would be amazing.
Acquiring/sharing data sounds good, but I am not personally planning to look at it before Sunday. Maybe someone else wishes to do this, and to bring it to the discussion? Relatedly, there are other good points here that I probably won’t manage to personally go into—maybe Oliver does, but if not, maybe we can have a discussion format that allows other people to also give short spiels/etc., and maybe if some are motivated to share such they can come with these? I am not sure what Ray and Oliver have planned.
In terms of concrete measures for lower-covid-risk school reopening, I think it’s worth underlining massive use of HEPA filters. They are not that expensive, and should reduce aerosol density and thereby covid risk rather a lot. Worth trumpeting also for grocery stores, pharmacies, offices, etc.
I like your overall point more though, and feel a bit innane commenting about something else without responding to that.
The big thing is that a clear majority thinks schools should not wait for teachers to get vaccinated, let alone for students to get vaccinated, before reopening.
Wait, how do you get this? Am I misreading the chart above this sentence? It looks to me like in the poll you quote, 55% thinks schools should wait, which seems in contradiction with your sentence.
Sometimes unconscious/visceral fears are kind of what I thought they were:
If I look down from a tall height, I experience: [vertigo, a slight increase in my heart rate, some similar change in my breathing, a slight increase in my tendency to “freeze” my muscles, a shift in my attention toward the height/downness, etc.].
I might’ve thought this response was a consciously chosen strategy for not falling. Except that it also occurs when I’m walking on a glass floor in a well-engineered building I trust. Still, I conceptualize it as something like “in-built fear of heights; designed to prevent falling but based partly on a bunch of visceral cues that persist even when my conscious mind knows I won’t fall”. My lead “rationalization” for the response (“I’m breathing more shallowly and refusing to walk to the edge of the cliff with you because I don’t want to fall off the cliff”) is at least partly post-hoc, and is causally downstream of a more visceral reaction… that is basically also evolved for not falling off cliffs.
And sometimes unconscious/visceral reactions are just not at all what I thought they were:
Some years ago, I was waiting impatiently for a friend to leave an event we’d both been at, so that I could “go home to do my important work with its urgent deadline.” Then I took a better look at the feeling, and realized that I was actually cold and in need of a washroom. I found a washroom and some warmth, and felt suddenly at ease and untroubled about my task. I’ve related differently to such feelings ever since.
Similarly, some years ago I was hanging out with my cousins around a holiday, and I felt “bored” (as I saw it) “because” they weren’t talking about anything interesting, and I couldn’t share with them my thoughts about anything interesting, such as AI risk. And then we switched board games, and I started laughing more, and I realized that my previous “bored” has actually been made of “socially uncomfortable”. And I got a bit better at identifying the “socially uncomfortable” visceral response, and calling it by that name instead of some other.
The interesting thing for me about the cashier incident above, was the rationalizations my brain produced around the “don’t share my views” impulse in my cashier situation seemed pretty transparently unsound (in my situation, in contrast to e.g. Oliver’s similar instance in-thread, or any number of previous times when I had more plausible … rationalizations? reasons? for not breaking an Overton window). If I was concerned about the cashier’s welfare: yes, she might find it uncomfortable to see our views, but “withholding practically relevant information from someone specifically asking about it” does not seem kind. If I was concerned about causing some sort of social drama that might do me harm: it didn’t seem like there was much plausible harm. (The risk that she might object to selling me the groceries did not occur to me, I think correctly. The 24-hour Safeway just outside of Berkeley may have had different customers than the downtown Berkeley Trader Joe’s that Oliver and Ben Pace mention. She seemed peaceable/stable/normal. I was with my husband, which is probably safer than alone. Etc.) There probably was still more social drama in telling the truth vs in evading, but its absolute amount seemed pretty small, to the point where the normal ratios at which I try to buy [human decency / communication / being helpful and honest] vs [avoiding harm to myself] seemed to pretty clearly favor talking.
So it was more a glass floor situation, than an actually being near a cliff situation. Useful for elucidating things. And I’m not yet sure what the analog of “avoid falling” is, that this reaction is actually triggered by cues of. Is there a pretty in-built visceral thing for “don’t break Overton windows”, that is quasi-independent of conscious knowledge that you’re safe? Is its true name “don’t break Overton windows”, or something else? What’s up with the way the impulse in me oscilated between selfish rationalizations (“she might harm me”) and morality-related rationalizations (“it’s wrong to upset people”)?
Totally. I was not AFAICT worried at the time about limited supply buying, or not very worried; the Safeway we were getting things from did not seem out of much and I hadn’t heard people complain about shortages/buying yet as far as I can recall.
The feeling of breaking an Overton window
To be honest I am not sure what exactly is being advised.
I am basically advising that you treat the concept of PR, and the word “PR”, the way you would treat a skilled but incredibly sleazy used car salesman. You may sometimes wish to deal with him anyway, if you can’t practically locate any other way to buy a car. But you’ll want to be very very alert to what’s being slipped into “your” “beliefs”, while you do so.
Sort of like if you were using a concept from Scientology to navigate a personal psychological issue.
Do you think trying to be ‘honorable’ will suffice to avoid bad outcomes?
I think that attention to “honor”, “reputation”, “brand”, etc. will get us most but not all of what we might hope for from PR, and including some things that PR itself won’t give, such as some kinds of longer-term freedom, grounding, and ability to think.
I would advise using this concept first (just, very simply, substituting the word “reputational concerns” for “PR concerns” in conversations, and seeing where this substitution gets you).
I don’t think it’ll do everything PR would do. And I’m not saying you should never care about the residual (although I am saying that the sleazy car salesman may have tricked us into sometimes thinking the residual matters more than it does).
It’s much easier to resolve disagreements about what counts as good PR.
I mostly disagree. I mean, maybe this applies in comparison to “honor” (not sure), but I don’t think it applies in comparison to “reputation” in many of the relevant senses. A person or company could reasonably wish to maintain a reputation as a maker of solid products that don’t break, or as a reliable fact-checker, or some other such specific standard. And can reasonably resolve internal disagreements about what is and isn’t likely to maintain this reputation.
If it was actually easy to resolve disagreements about PR, I suspect we wouldn’t be so spooked by it, or so prone to deferring to outside “PR consultants”.
I… my thoughts aren’t coherent enough here to let me know how to write a short comment, so I’m gonna write a long one, while noting aloud that this is a bad sign about my models here.
But: it isn’t just a matter of deferring to polls. Partly because with “bad PR” or scandals, there’s a dynamicness to the mob. It isn’t about peoples’ fixed standards or comparisons, that you could get by consulting polls. (Or just by consulting a friend or two, the way you probably do when you are faced with normal ethical questions and you want help remembering what the usual standards are.) It’s some spooky other thing, involving dynamics that evolve, and experts that you pay to be a bit distanced somehow from your not-knowing and to be able to tell other people that of course you consulted an expert so that they won’t shun you after the whole thing explodes.
It seems like it’s easier for organizations to coordinate around PR
So, that does seem true, at least in the sense that lots of organizations and groups talk about “PR”; so there’s some tautological sense in which it’s gotta be easier for organizations to somehow end up talking about that. (Lately. It wasn’t so in past centuries, FWIW. Possibly they just didn’t know how.)
But I am a bit unclear on why.
One hypothesis that matches my own introspection, at least for small- to medium-sized organizations of the sort I’ve been involved in, is that we attend to PR because it’s somehow part of the received “everyone knows you should pay attention to PR” morality that was imparted to us, right next to “don’t drink and drive” and “get a college degree” and “remember to feel a sense of ‘wow’ if somebody mentions Harvard”. And not because of inside-view/directly-perceivable advantages to attending to PR (vs reputation/brand/honor).
Of course, this just kicks the can down the road — why would this morality have been imparted to us? I’m honestly not sure. I don’t trust it. I do not personally notice myself or others making any worse decisions when we instead attend to “reputation”.
Thanks; I find this comment helpful and interesting, like part of a puzzle.
Trying to advance AI in hopes that your team will get all the way to AGI first, and in hopes that you’ll also somehow solve alignment at the same time. Backfires if your AI-advancing ideas leak, especially if you don’t actually manage to be first. Backfires worst in worlds where you came closest to succeeding by actually making tons of AI progress. Reminds me of “shooting the moon” in the card game Hearts that way.
Sometimes I wonder whether the whole current AI safety movement is net harmful via prompting many safety-concerned people and groups to attempt AI capabilities research with unlikely “shoot-the-moon”-style payoffs in mind.