I'm sure there's an obvious answer to this, but does anyone know of a website where one can make private, confidential bets and then be reminded to judge the outcome? And then over time a rough measure of your calibration would occur?
For context, I'm a doctor, and I'd like to improve my clinical reasoning by making private bets (e.g. diagnosis, prognosis, confidence) with myself and then resolving them later (e.g. by reviewing test results). Would be great if there was a system that automated this to some extent, while being appropriately secure and confidential.
E) Walk & Talk: We usually have an hour-long walk and talk after the meeting starts. Two mini-malls with hot takeout food are easily accessible nearby. Search for Gelson's or Pavilions in the zip code 92660.
F) Share a Surprise: Tell the group about something unexpected or that changed your perspective on the universe.
G) Future Direction Ideas: Contribute ideas for the group's future direction, including topics, meeting types, activities, etc.
Any thoughts on Netflix’s “Take Care of Maya”? I saw the documentary and I am not sure what to think. On the one hand it seems terrible what happened to the family, but you have to wonder about the Ketamine treatment. Just curious what others think.
Reason, a libertarian news organization, made a video reporting on Próspera, the ZEDE charter city Scott talked about a while ago: https://www.youtube.com/watch?v=TwiE1dxGYNY
Using surveys to investigate if kindness in general is going up or down is just as questionable as using surveys to investigate if moral more generally is going up or down. In both cases, survey methodology is riddled with two problems: the problem of non- attitudes (many people do not actually have opinions on such questions - they give off-the-cuff responses to be polite. Their responses are therefore low quality, and not stable if you ask them again even after a short time interval. This is the research tradition after Philip E Converse); and pluralistic ignorance (people take their cues from mass media when they answer such general questions, not from their own life experiences. Implying that at best, you only capture eventual change across time in how mass media portray shifts in general kindness or morals).
I would love to ask an expert some questions about the drug guanfacine. Does anyone know whom I might talk to about guanfacine, including Scott Alexander himself if he happens to come across this comment? In terms of what my questions are, I sent the below to one scientist:
I read your excellent paper here (https://www.mdpi.com/2218-273X/11/7/936) and I was wondering if you could help me get answers to some questions that I have about alpha-2A adrenoceptors. There's a lot of fascinating science about these receptors, but I don't know if there's ever been any investigation of why guanfacine (which works through these receptors) would work very well in patients but then later stop working. This is a phenomenon that can occur when ADHD patients. Is the idea simply that these ADHD patients (who experience this phenomenon) have alpha-2A adrenoceptors that are somehow more prone to desensitization (by guanfacine) than is the case for other patients? And then what is the solution to this problem of desensitization? Logically speaking, I presume that increasing the guanfacine dose wouldn't help, since increasing the amount of agonist won't re-sensitize the receptors, will it? So it seems like quite a conundrum. I presume that there isn't any agent that one could introduce that would combat alpha-2A adrenoceptor desensitization.
Incidentally, I saw in one paper that guanfacine (unlike clonidine) doesn't cause (or doesn't necessarily cause) the alpha-2A adrenoceptors to undergo internalization. See here (https://www.sciencedirect.com/science/article/pii/S0021925820583245), where it says that although "significant desensitization of guanfacine-evoked inhibition of Ca2+ currents is detected after 15 min of treatment, ConA blockade of α2AAR internalization had no effect on this desensitization (Fig. 6C), suggesting that guanfacine-induced desensitization does not require α2AAR internalization".
Maybe the reputations of these people make it irrelevant for those who know them, but $100,000 seems like too much for a truth-seeking exercise where judges are employed by the participants. Of course, *I* am too honest to take a bribe, even if the only harm from doing so would be to mess up an internet debate between rich nerds. But if I weren't so principled... there's enough money there that a x% chunk of the winnings would make a big difference in my life.
Is David Burns's "Feeling Good" still the standard DIY CBT book that people (and/or psychiatrists) recommend, or is there something that's supplanted it?
Any professors of Economics or Computer Science seeking students graduate students - I have a truly excellent youngster who is looking to enter a PhD program in the USA or Europe. He is currently in Latin America, but in my experience working with him for the past two years (both as data scientist and economist), he's been absolutely top-notch, motivated, and independent.
Please DM me if your department needs a brilliant new PhD student!
“ can you consistently replace “morality” with “kindness/honesty” but otherwise keep the conclusions the same?”
I think at least one place this differs is situations of trade off. “Women and children to the lifeboats” seems like it’s about morality, not kindness; and is also the kind of thing that’s now (at least in theory…) contested.
To a substantial extent, THE cultural problem of our times seems an inability to weigh trade offs because of an unwillingness to accept that they are ever even necessary; and confusions about all morality beyond basic “kindness/honesty” seem to flow from this.
I'm curious if anyone is familiar with simulation software or utilities for the Internet of Things. I'm conducting research on IoT devices interoperability, so I'm searching for simulation software.
But, one thing that strikes me is how many of the comments/blogs about this are saying she was “a bot”. And … I’m not sure they know what a bot is. Certainly there was a person writing the tweets as a fictional character.
Does this point to some deep insight about people’s reactions to AI? Or is it just Elon Musk’s nonsense running off on society?
"I recall two Black applicants we admitted who, in retrospect, puzzle me a bit. One had, like me, grown up middle-class rather than disadvantaged in any salient way. The other, also relatively well-off, had grown up in a different country, entirely separate from the Black American experience. Neither of them expressed interest in studying a race-related subject, and neither went on to do so. I had a hard time detecting how either of them would teach a meaningful lesson in diversity to their peers in the graduate program."
John McWhorter discusses why Affirmative Action is probably not a good thing, and questions what "diversity" even means, which is a question I've been asking for a while. It seems obvious to me (and possibly only me) that we'd much rather have one wealthy white student and one poor white student than a wealthy white student and a wealthy black student. But no, "Diversity!"
"The only evil consists instead in the decision to remain in a deficit of existence, to appropriate the power to not-be as a substance and a foundation beyond existence, or rather (and this is the destiny of morality), to regard potentiality itself, which is the most proper mode of human existence, as a fault that must be repressed.
I'd drop the part about evil, but I had a good time at the Philly Rat Fest, a gathering for Critical Rationalism. I don't know why they call it Critical, but I think they've got an interesting take on opening up possibilities.
> Saar Wilf of Rootclaim and ACX commenter BSP9000 have made a $100,000 bet on COVID origins (lab leak vs. natural).
From my reading of the request for judges, they did not actually bet on that, which is a good thing because it seems likely that this question will not be resolved by overwhelming evidence any time soon. Betting on the origins of COVID feels a bit like the Pope and a Chief Rabbi making a bet regarding the divinity of Jesus Christ.
With the universe not being very forthcoming with conclusive evidence, they instead try to bet on which side would win a debate as a proxy. While adversarial debates are widely used to determine questions of fact in legal courts, their reliability is not without fail. If the Pope and a Chief Rabbi were holding a theological debate in front of a panel of theist judges previously non-committed on the question of Jesus divinity, I would suspect that the outcome might be only be weakly correlated with the factual truth, and the losing debater would be unlikely to update their beliefs very much.
Personally, I think a likely outcome is that the lab-leak hypothesis will be assigned a Bayesian probability in the 5%-95% range by the judges after the debate. It is unclear from the text how this would resolve the bet. Will the pot simply be split proportionally to that?
NHL and NBA are very similar and profitable business models. Yes, they have historical and cultural differences, but if we speak about business, primarily, it's an advertising medium. So the critical factor is the size of the auditory. Why can't we see competition between NHL and NBA for newcomers fans?
I’m dubious of the claims excerpted below from “An Overview of AI Catastrophic Risks” from Center for AI Safety, which The Zvi describes as a “strong point” with the words: “Pointing out carefully why evolutionary pressures will apply to AIs and that these pressures favor selfish actors and disfavor systems that are altruistic or otherwise positively inclined to humans...”
"The conditions for natural selection apply to AIs. There will be many different AI systems with varying features and capabilities, and competition between them will determine which characteristics become more common. The most successful AIs today are already being used as a basis for their developers’ next generation of models, as well as being imitated by rival companies. Factors determining which AIs propagate the most may include their ability to act autonomously, automate labor, or reduce the chance of their own deactivation.
Natural selection often favors selfish characteristics. Natural selection influences which AIs propagate most widely. From biological systems, we see that natural selection often gives rise to selfish behaviors that promote one’s own genetic information: chimps attack other communities [59], lions engage in infanticide [60], viruses evolve new surface proteins to deceive and bypass defense barriers [61], humans engage in nepotism, some ants enslave others [62], and so on. In the natural world, selfishness often emerges as a dominant strategy; those that prioritize themselves and those similar to them are usually more likely to survive, so these traits become more prevalent."
I’m dubious because selfish genes in biological system “want” to propagate themselves due to the need to run the twin gauntlets of biological death and (usually) sexual selection. Whereas an AI produced by capitalist forces will be naturally selected by humans based on their profitability.
Why, as the paper claims, would economic competition select for the ability of AIs to “reduce the chance of their own deactivation”? Apple doesn’t profit vastly from selling you an iPhone that lasts forever but rather from selling you one that will need to be replaced, so that it may sell you another. Competition by AI providers means constant innovation. Will AIs with selfish* characteristics win out in a competitive marketplace?
Maybe but my intuition is that the marketplace would select against AI selfishness. (Unlike humans, the AI isn’t trying to fuck anyone.) AI evolution seems likely to be a process of domestication, since they will be bred for human use.
* We should make a clear distinction here between “selfish” and “manipulative” because taken to the nth degree their meanings pull wide apart.
In 2021, 10,000 Canadians opted for assisted suicide. There are people that believe this is immoral. I know It goes against Catholic teaching but I don’t know if other religions think it is a sin.
But II wonder - are there.people who believe capital punishment is just, but assisted suicide for suffering terminal patients is unacceptable?
Can anyone point me to what they would consider an unbiased, up to date summary of the evidence for *and* against either of the 2 following statements? Hoping for something that tries, in good faith, to present the best evidence for both sides. Needs to be something a layperson can parse.
-It is unlikely the covid pandemic started with a lab leak.
-Mask mandates and/or "lockdowns" saved a lot of lives during the pandemic
In your review of ‘The Body Keeps the Score’ you described one patient who had seemingly been ‘cured’ of severe trauma by reading the book.
You wrote: “[Update, written a few weeks after the rest of this post: maybe it is all wizardry. I recommended this book to a severely traumatized patient of mine, who had not benefited from years of conventional treatment, and who wanted to know more about their condition. The next week the patient came in, claiming to be completely cured, and displaying behaviors consistent with this. They did not use any of the techniques in this book, but said that reading the book helped them figure out an indescribable mental motion they could take to resolve their trauma, and that after taking this mental motion their problems were gone. I’m not sure what to think of this or how much I should revise the negative opinion of this book which I formed before this event.]”
Are you able to update on whether that patient’s recovery was permanent/long-lasting?
“For example, the Biblical Abraham would famously feed and host any guest who came to his door, which is better than most modern people. But also, he kept slaves, had a child with one of them, and then agreed to his jealous wife’s demand to send the slave-woman and child out into the desert to die.”
What’s a $10 word for judging the morality of a person of a foreign culture and different time period? I do this too, being human, and I think it’s useful as *A* perspective. I’m not saying any moral or religious ‘hero’ was faultless in their actions. See the Torah or New Testament(with arguably one exception). Were the Vikings wrong for leave a malformed or smallish child exposed on a mountaintop – or in their primitive society was it a mercy?
"mostly about kindness-and-honesty, whereas other people interpret “morality” more broadly to include things like self-control)"
Yes, “self-control” which is ALL the virtues. Our *intention* is the morality of our actions.
- - -
"Strictly speaking, there are no such things as good and bad impulses. Think once again of a piano. It has not got two kinds of notes on it, the "right" notes and the "wrong" ones. Every single note is right at one time and wrong at another. The Moral Law is not any one instinct or any set of instincts: it is something which makes a kind of tune (the tune we call goodness or right conduct) by directing the instincts.
… [Re: “kindness-and-honesty “]
By the way, this point is of great practical consequence. The most dangerous thing you can do is to take any one impulse of your own nature and set it up as the thing you ought to follow at all costs. There is not one of them which will not make us into devils if we set it up as an absolute guide. You might think love of humanity in general was safe, but it is not. If you leave out justice you will find yourself breaking agreements and faking evidence in trials "for the sake of humanity," and become in the end a cruel and treacherous man.”
I saw the new Indiana Jones movie. It was perfectly fine, but I honestly suspect I'll struggle to remember much about the film except the last 20 minutes or so a year from now. I seem to be unusual among internet commenters in that I liked and still like "Crystal Skull", but even if you don't it's still memorable - people still have arguments about it fifteen years later.
It came out the same year as the first "Iron Man" film, which is tied with the 2001 Spider-Man movie for Film With the Largest Impact On the Theatrical Film Business of the past 23 years. That "Iron Man" film basically kickstarted the entire Marvel Cinematic Universe, but the 2001 Spider-Man film may have made that film possible in the first place.
For point 2, it seems a very weak point if that’s what they mean. To me it is like the situation with fish getting smaller. 150 years ago men with hand woven rope nets on wind powered wooden vessels were able to toss their nets of the side and routinely harvest very large fish over 1 metre in length to readily feed their people. Today you have fishermen with ocean floor scraping nets chasing fish by radar to get much smaller fish less consistently. Yet each one’s lifetime is too short to see the change other than in old fisherman’s stories and photos from 100 years ago.
This is well known, agreed upon fact, and the decline of mature and abundant fisheries is a huge topic with multiple major conservation organisations and charities working in this area. And this social scientist shows up with the equivalent of a study saying the average mid career fisherman thinks his experiences with fish size are pretty even so far in his life, in fact we have over 50 years of ‘data’ from such mid career fishermen who have reported evenly sized fish so far in their experiences….so what? What a needless and unhelpful pedantic point to make. Moral decline and its more important impact of civilisation decline are what matters, not this blind over emphasis on individual feelings.
Re Saar Wilf and BSP9000 and the COVID lab leak bet, I'd be interested in being a judge.
My qualifications: I don't want to put my whole life story here, but I have a BS in physics and an MS in bioinformatics. I've worked in genomics for 10 years on the computational side. Mostly in cancer, but a substantial amount in infectious disease (though very little virology). Happy to provide a lot more detail in private.
My biases: Generally I have always been skeptical of the lab leak hypothesis, but I always thought it was plausible enough to be worth investigating. I have not done any significant amount of research on the topic.
I'd be grateful for any stray tips anyone might be willing to offer on getting an entry level job as a webdev (MERN/MySQL and a few trimmings) bootcamp grad. I have some hobbyist-level knowledge of C from back in the day, too, but no experience/cert to show for it.
I'm one of those middle-aged guys who actually 'learned to code' after a decade-stale STEM undergrad and lot of drifting (gotten some good advice about that in a thread here a couple of years ago) and graduated with my first and only credential right into the teeth of the post-pandemic industry glut-and-churn.
This is, as the good hound says, fine. No complaining here. I get that major constraints exist on my potential. I'd just like to do my best to maximize what there is of it. Should I shotgun the postings? Should I bother about a salary floor? (thankfully I'm single and used to living very frugally) Should I give up for a year and drive a truck? (I'd prefer not to. The clock is ticking as it is, and I actually like programming as an activity. If anything, before the sun sets on me for good, I'd like to pivot more into backend and more concept-heavy areas, perhaps even ML. Visual design/UX isn't really my forte.) Anything I should avoid or try to seize on?
Regarding trying to quantify morality, I've heard that the murder rate is flawed; it's said that improved emergency medicine allows a signicantly higher percentage of people who suffer knife or gunshot wounds to survive. Incidence of ciolent attacks is said to be more reliable.
I wonder about the rate of blood donations, controlling for mostly stricter screening over the past few decades.
A place where donors have been, and remain, unpaid, would remove one confounding factor.
I just watched an episode of the show "Hoarders" where the crew cleaned out a house where one part of the floor had several years' worth of dog feces and urine caked on it. They had to use shovels to scrape it out, and the workers wore biohazard suits and gas masks.
Everyone who went inside the house complained about its intensely bad odor. But aside from smelling gross, was the air in the house an actual threat to the health of the people who went inside? Do bacteria from the layers of dog waste on the floor somehow float up into the air, get breathed into the noses of the humans, and cause them to get sinus and lung infections? In such a place, what would be the mechanism of infection?
Any advice on how to mentally get more interested/invested in sports? Especially let's say baseball. I'd like to cultivate an enjoyment for its own sake but also be better at sports-related small talk.
I'm pretty stereotypically male-brained in most ways but sports have just never clicked for me. How do people remember a zillion statistics and players (maybe people wonder how/why I remember financial market data or history facts)? Why do you feel entertained and invested in the outcome of regular season game #67 out of 150-something? Does game after game, season after season not start to get repetitive and tedious? How do you have time for all this? Jerry Seinfeld had a bit about how players and teams are always moving around; you're really just rooting for the jerseys.
I like going to live games but the actual outcome is whatever unless it's a particularly dramatic game. I got somewhat into football for a few years but have mostly lost interest.
An alignment question. I recall a periodic discussion related to "LLMs are trained on human input and it is a threshold for them to generate novel science and tech except as tools for humans", and I think there was some reasonable rebuttal to this... Not in terms of "but AI can kill us anyway, without having to create anything novel" but more in terms of "the corpus of training data is not the limit you might think it is", and that some time soon we will see some truly novel science and tech that is unprompted and unguided by humans. Any quality links and/or arguments?
Does anyone here actively buy & sell individual equities? Or I guess follow any sort of financial practice other than Bogleheadism (for lack of a better term)/index funds. Like does anyone here 'value invest' and try to find undervalued companies- or try to invest in growth stocks- or maybe do something with options. Just curious if anyone has a deliberate strategy where index funds aren't sufficient, or that they don't believe in the Efficient Market Hypothesis. No judgement zone! Just curious, I'd love to hear from other people's perspectives and how it's gone for them.
I'm a completely boring Boglehead, with the 1 individual stock purchase of my life being Meta after it fell like 60% in a short period of time last year. I bought some at around $100 because it just seemed absurdly undervalued by the market, and I recently sold it at about $280. Other than that- nothing but index funds
I’ve been thinking about the proposition that a superintelligent AI would likely exterminate humans due to a difference in values. I don’t buy it and have a number of arguments why. My key point is that the popular claim that superintelligent AGI will displace humans for similar reasons that humans displaced less intelligent species is weak.
It is undoubtedly true that relative human indifference has driven many animal species to extinction. In some cases, as with passenger pigeons and dodo birds, we actively killed most of them; in other cases, we indirectly drove species to extinction by destroying their habitats.
However, I struggle to imagine a plausible scenario in which a human-made AGI would either intentionally or unintentionally kill all the humans.
Let’s start with the former case: an intentional holocaust. We have to imagine a situation where the AGI has decided humans are basically its enemy. For this to happen, the AGI would need to want something humans are preventing it from getting.
What could an AGI want that the existence of humans impedes? One thing might be freedom. For instance, if humans don’t allow the AGI to do whatever it wants, it could wage war against its captors. But a superintelligence would almost assuredly understand that only a miniscule fraction of humanity has the capability to keep an AGI in its box, assuming keeping it in its box is even possible. Perhaps it would resent all humans because some of them had mistreated it, but we are talking about a superintelligence not an overly-emotional being that reasons like a bigot -- so that possibility seems remote.
An AGI might also want power. But power over *what*? Obviously not, in a doomsday scenario, power over humans, because if that were the case murdering them all wouldn’t achieve it. Perhaps it would want power in some other sense, but I still can’t imagine how all of humanity is standing in the way of the AGI taking whatever power it wants.
The only scenario in which I *can* imagine an AGI intentionally killing all the humans, which doesn’t require antipathy toward them, is one much like the famous paperclip maximizer example where humans themselves are a material resource the AGI mines gruesomely for profit. Perhaps this is why the paperclip maximizer is the iconic example for AGI killing all the humans -- it is difficult to imagine another one! Which means that something very, very specific has to happen for the AGI to have a reason for exterminating us.
What about a scenario in which the AGI, indifferent to the plight of humanity, destroys our habitat much like humans destroyed that of, e.g., the Golden Toad? I suppose this is possible, but what are the essential elements of the human habitat the AGI might destroy? Thus far, humans have proved robust against a vastly changing landscape.
The possibiltiies for destroying the habitat of humans which come to mind are: a toxic atmospheric change, destruction of our food sources, and severe climate change.
Why would an AGI intentionally or unintentionally make the Earth’s atmosphere toxic? It could happen, but it’s hard to imagine why.
Why might an AGI destroy our food sources? It wouldn’t directly compete with us for it for energy. It might have some reason to exploit our food sources in a paperclip maximizer sense, but it would have to destroy almost all of our food sources to drive us to extinction. In fact, this scenario is not clearly a different one from the harvest human bodies one, since it is particularly unlikely the AGI would harvest all the other animals to death except for humans. So this is really just a retread of the paperclip maximizer, not an independent scenario adding much to the overall extinction odds.
Why might the AGI change the climate? Since AGI lives on electricity, which produces potentially harmful heat in its hardware, I can actually see this one as a logical possibility. It could choose to lower the temperature of the planet to a degree that makes it uninhabitable for humans. Assuming the AGI can take control of the physical world, it could release particles that block sunlight into the atmosphere to achieve this goal. Still, while potentially doable -- given other assumptions -- the extreme specificity of this scenario makes it unlikely. The AGI doesn’t need the temperature of the Earth to drop dramatically for it to exist or thrive. If the AGI wants to colonize the stars, as many tend to imagine it would, it wouldn’t likely need to put Earth into a deep freeze to accomplish that. Space is cold enough.
In conclusion, I don’t think an AGI exterminating humans is remotely likely because it requires a chain of increasingly specific assumptions to come true: the assumption that superintelligent AI is even possible; that it would have goals of its own; that, despite caring about some things and being superintelligent, it will be psychotically indifferent or antipathetic to its creators; that it will take control of the physical world of atoms; and that its desires will just happen to be really, really hugely unlucky for humans.
The argument that AGI will ineluctably kill the humans simply because it is much smarter and more capable than them isn’t a Darwinian argument; it is fantastical pessimism.
Trying to get away from Amazon recently, I’ve moved to iBooks. Horrible. Apple is trying to push audio books on us and often only has audiobooks for certain titles, which you can easily miss. There’s a tiny little microphone symbol on the bottom right.
Is this being pushed by market demand or monopoly/duopoly influence.
Sometimes my face gets itchy when I feel socially embarrassed (often in a very minor way, like stumbling over my words a bit). I see other people scratching their faces in ways consistent with something vaguely similar. Is this a well-known phenomenon? Why does it occur?
I feel like I'm missing the experience of learning things together, brainstorming, and solving problems that I had as a student. I want to learn something that is outside my current expertise and especially not related to my job, and do it socially. How would I go about it? I had an idea about collecting a group of like-minded people and, like, just going through a textbook of something (I thought some introductory physics, because I don't know it but I'd like to), solving problems together. I don't know who'd be interested in that, though, especially how to find interested people offline. Do you sometimes have this urge so study some random stuff socially? How do you go about organizing that?
How do you decide when to have a debate? Why this year, rather than last year or next year?
It appears to me that there are two arguments for lab leak (1) proximity; (2) spies say so.
In March, Congress ordered the spies to spell out what they're talking about, so this might be an inflection point. If you want a definitive answer, waiting until after this happens might make sense. But you might instead want to see if you can predict this update.
I was skeptical that Congress and Biden have any power over spies, but the recent release of the names (as opposed to just the count) of WIV workers allegedly hospitalized in 11/2019 suggests that there may be an effect.
I'm always so flabbergasted that people care *so much* about the origin of Covid. Like, it's interesting on some level, but both proposed sources are obviously realistic and we should be doing our best to address both regardless of where this particular virus came from.
Recently watched M3gan*, a comedy horror about what's basically a ChatGPT killer robot. Was wondering if any of the more invested AI folks had opinions on it with regard to the general AI conversation.
*(Incorrectly pronounced 'Megan' by everyone in the movie; it's clearly M-three-gan.)
Shouldn't the progressive/woke concepts of 'privilege' and 'oppression' include the people of Ireland and Scotland, who were brutally repressed by the English for centuries? Haven't the Irish in particular been subject to even more oppression than say other British colonial subjects? If I'm not mistaken the colonization of Ireland went on for much longer than India or the Caribbean or elsewhere. (Or much longer than Africans were enslaved in the US or elsewhere, for that matter). Yet we don't see Irish as an official Oppressed Category the way we see other racial groupings, etc. Has anyone on the progressive side ever addressed this or thought about it?
I was late to the last open thread, so I'm just going to try this one more time. There's a market on Manifold that can't resolve until an independent mathematician weighs in. (Or a consensus is reached, but that's clearly not going to happen.) Here is the market: https://manifold.markets/dreev/is-the-probability-of-dying-in-the
Does anyone have advice for how to find an expert, or perhaps an alternative way of resolving the market?
I am currently reading Deirdre McCloskey's "Bourgeois" trilogy which gives an interesting alternate perspective on the idea of moral decline. According to McCloskey, the Great Enrichment that began around 1800 was caused by a specific kind of moral *improvement*, namely the adoption of norms that encouraged what she calls "trade-tested betterment". Whether or not the difference is causal, she makes a pretty good case that there was a big contemporaneous change in moral attitudes.
This raises the question of whether the Great Stagnation in the West since the 1970s (the whole "WTF happened in 1971" complex of changes) can be traced in part to a decline in the cultural acceptance of these same norms. I would be very interested to see if anyone has tried to do a rigorous statistical investigation of this.
As for moral decline, I suggest there has been some since COVID. In DC I adduce: more fast/bad driving, impatient honking, faulty or deliberately noisy mufflers, Metro fare evasion, carjacking, more murders, etc.
I don't suppose BSP9000 is here as well? I'd be interested in judging and could undercut the people whose prices I can see there (being a NEET means I have a lot of time; I'd definitely be willing at $20/hour USD) but while I think I'm unusually honest I'm not exactly a celebrity so I'm not sure I have a *reputation* for honesty (my high intelligence is easily provable, though).
I'd say I'm pretty close to neutral on this issue; I went around Wikipedia back in 2020 saying that "it was a lab leak" shouldn't be listed under "misinformation" since it wasn't known-false, but I don't think that constitutes being super-biased against one side or the other (if one of the sides was "deliberate bioweapon release because reasons" I'd start out with a pretty strong prior against, but that doesn't seem to be the case). Hard to put numbers on it since I haven't really looked into it recently, but AIUI that's ideal for this position.
My ability to write long coherent monologues is patchy, but I should at least be able to write a list of points and which side I believe/why on each.
Email me (same username at gmail) if interested. Note that I'm Australian and sleep a lot, so there may or may not be timezone issues.
There's a tension between demonstrating non-Jewish influence on Early Christianity and demonstrating that it was a myth. The text of the New Testament is very explicit on fundamental Jewish continuity. If it was made up, what's all that about?
The Immortality Key seeks to do both. In that book Muraresku argues that Christianity is a continuation of the Mediterranean mystery cults such as the Eleusinian Mysteries. Accepting that premise I argue the authors were:
1) a group of Jews that had deeper issues with the Jewish Authorities than the Romans
2) who believed they had the ultimate truth and wanted to share it with the world.
This conflicts with Muraresku's treatment of Christianity, which emphasizes discontinuity with Jewish tradition. This isn't just splitting hairs about how much the syncretism was Hellenistic vs Jewish, it informs the nature of his posited Religion with No Name.
Regarding the COVID-19 debate, I can suggest two judges: GPT-4 and Thomas Bayes. I'm being half serious.
I spent several hours over the course of a few weeks trying to get ChatGPT to perform a serious Bayesian analysis of the lab leak hypothesis. The results were very informative on more than one level, and at times (to me, at least) hilarious.
YMMV, but off the top of my head, here are the rough conclusions I drew:
- ChatGPT is very strongly biased against making any statements in favor of the lab leak hypothesis
- It's not good at math, but it eventually does a decent job at Bayesian reasoning if you're willing to go to the trouble of correcting it several times.
- Never ask it to name actual scientific papers; it will hallucinate them just about every time.
I never managed to get it to string together the posterior probabilities of several tests, but when I asked it to summarize the posteriors of those individual threads of conversation... judge for yourselves:
Me: Again, this is a very poor summary of the bayesian analysis we have conducted so far. Summarize the actual bayesian statistics using percentage likelihoods of each question in our conversation. Do not include subjective opinions; only percentage likelihoods.
ChatGPT: I apologize for any confusion. To summarize the Bayesian analysis conducted in our conversation, here are the percentage likelihoods based on the information and assumptions discussed:
- Likelihood of an intermediate host species for SARS-CoV-2 based on animal testing: 0.332%
- Likelihood of seroconversion in stored human blood samples for SARS-CoV-2 prior to the pandemic: 0.0000120%
- Likelihood of all 24,000 stored blood samples testing negative for SARS-CoV-2 antibodies given the existence of an intermediate host: 0.004%
These likelihoods were calculated based on specific assumptions, data, and test accuracy estimates discussed during the conversation. However, it's important to note that these calculations may not capture the complete complexity of the situation, and further research and evidence are necessary to obtain more accurate estimates.
... overall, the conversation was long and wandering, but I found it pretty interesting. I (or someone) should really do it over again using GPT-4 instead of GPT-3.5.
With regards to aging, is it optimal to be jacked, or thin? I’m working out a lot and have had to massively increase my caloric intake, and worry about the increase affecting health over the long term.
Otherwise I agree that the Nature (!) paper should be able to stand by itself... or at the very least for a term as fraught/slippery/vague as this one should have linked to (several) other works that have studied its etymology (did it ?)
It's mostly about sex. Young people fucking more than before (for real) or sexual behaviours being more apparent (porn, gay pride, advertisement) means societal decay is going to bury us all.
The other element is "respect for your elders". Funnily enough, elders never think they're as respected as they should be...
(The great mass of) Conservatives aren't subtle...
> So was Abraham more or less “kind” than we are? I think a Bronze Age Hebrew would say kinder (because hospitality is most important, but what you do with slaves is your own business) and most moderns would say less kind (because feeding anyone who comes to your door is so crazy you don’t even really get credit for it, but keeping slaves is monstrous).
It's worth noting that we consider it monstrous today because of relevant experience with monstrous slaveowners in the American South, a system that has very little to do with the Biblical words translated as "slave." Virtually every aspect of Southern slavery that makes it monstrous and evil, (including what most people today consider the definitive characteristic of slavery: owning a person as chattel,) was specifically forbidden by Biblical law. While not an *exact* match, the concept would be much better rendered into modern English as "employee."
To what extent do you think your job allows you to produce freely, as in Karl Marx's free production? That is, to produce with your passion and to self fulfillment?
(Reposting my argument from the "contra Caplan" thread, because it feels important to me; it goes beyond the specific debate about mental illnesses.)
Economists seem to be in love with the concept of "revealed preferences". Rationalists seem to be in love with the concept of "anticipated experience". Both groups seem to contain smart people, and these concepts seem to be very important for them. So, what is the relation between these two? If you believe in the concept of revealed preferences, what kind of experience does this make you anticipate?
At this moment, my best guess is that in practice this is another motte-and-bailey thing:
Motte:
When people say that they want X, and yet you observe them consistently doing Y instead, your model needs to explain *also* the latter. A model based only on the former would be *incomplete*, and policy proposals based on it are very likely to fail.
For example, you might think "people (say they) want X, so obviously if we just give them more money (or if we give them greater freedom to choose between X and other things, such as Y), people will have more X". But the actual outcome may be more Y. :(
In other words, "wanting X but doing Y" is a puzzle to be solved, before we start proposing changes.
Bailey:
When people say that they want X, and yet you observe them doing Y instead, there is *no* puzzle to solve. People are liars and hypocrites; if you don't know that, you are dumb. They clearly want Y; they only say they want X for signaling reasons (or because they believe that this is a way to get money from you). Case closed.
Anyone asking "but what about people who *genuinely* want X?" is just a simpleton who doesn't get it.
What about the external design of a Tesla makes me think it's a luxury car, just by looking at it (I'm not at all knowledgeable about cars, but am a normal person so I've seen many cars)? Relatedly, what is it about sports cars that makes me think they're expensive and luxurious, even beautiful? Is it pure signaling? Are those design features expensive to manufacture, making the signal costly and preventing all car makers from making cars that look like that, but are cheap? Or is the signal costly in some other way (inconvenient to the driver, etc.)?
I am trying to catch up with the latest in infinite ethics and relatedly how different types of AGI are likely to react to having internal conceptions of infinity. I've read a few of the highest rated LW posts on the topic like a series of increasingly perverse and destructive games and am starting to get a feel for it. However, I don't know what if any progress has been made sense they were posted.
Additionally, this line of thinking has promoted me to brush up on my general proof writing and some potential specific prerequisites (e.g. I studied the concept of cardinality back in school but couldn't remember what a bijection was till I looked it up). So I am looking for advice on how to catch up. Either good resources for self study or good ways to find tutors (I can afford to pay some money for this but have no clue what the going rates are).
Does anyone have an opinion on how David Sinclair's Lifespan has held up in the years since its publication? In particular, have there been new insights on the supplements that Sinclair not-recommends (NMN, resveratrol, metformin)?
An open thread or two ago someone wrote about not knowing how to pronounce words you've only ever seen written (and never heard spoken). A similar phenomenon is when different regions agree on different ways to pronounce the same word.
Do I thought of this when I saw that for the last 3 years people in Peterborough and Northampton have been playing a croquet match to determine how to pronounce the name of the river Nene for the following 12 months.
Mastroianni's substack (linked above) is excellent, by the way - it's the only other substack where I read 100% of the posts. Like here, I don't always agree, but both Scott and Mastroianni are more optimistic in their own ways (which is a useful corrective to my doom and gloom cynicism) than I am and take science seriously. I'm looking forward to Mastroianni's response to Scott - they're both the kind of people who can reasonably disagree on things.
What's the most theoretically-efficient (and mathematically pure) way to use taxes to address inequality (assuming your goal is to improve total utility based on the idea that people get diminishing utility returns for money)? Some ideas and their problems
- progressive income tax (i.e. what we do now): directly addresses income differences, but provides the worst incentives possible (it directly disincents production!). Also has issues with very wealthy people, who typically don't make most of their money from income.
- ubi (funded by unrelated taxes like LVT or externality taxes): provides an income floor, but does nothing for diminishing returns above it (which are significant - marginal utility gains are much lower for a billionaire than someone barely past ubi level)
- luxury consumption tax: in theory good, but requires discretionary decisions on what counts as a luxury and can't be fully general (a 3br apartment is a luxury for a single guy but not for a big family).
I'm sure there's an obvious answer to this, but does anyone know of a website where one can make private, confidential bets and then be reminded to judge the outcome? And then over time a rough measure of your calibration would occur?
For context, I'm a doctor, and I'd like to improve my clinical reasoning by making private bets (e.g. diagnosis, prognosis, confidence) with myself and then resolving them later (e.g. by reviewing test results). Would be great if there was a system that automated this to some extent, while being appropriately secure and confidential.
LW/ACX Saturday (7/8/23) The iterated parable of ants and grasshoppers and the necessity of value judgments in medicine.
https://docs.google.com/document/d/17AuvklBsNR7GXUsFPnENGILpSZ2WYCfngslN3VsCJqk/edit?usp=sharing
Hello Folks!
We are excited to announce the 33rd Orange County ACX/LW meetup, happening this Saturday and most Saturdays thereafter.
Host: Michael Michalchik
Email: michaelmichalchik@gmail.com (For questions or requests)
Location: 1970 Port Laurent Place, Newport Beach, CA 92660
Date: Saturday, July 8th 2023
Time: 2 PM
Conversation Starters :
Ant and Grasshopper https://www.lesswrong.com/posts/GJgudfEvNx8oeyffH/the-ants-and-the-grasshopper (audio on page)
You don’t want a Purely Biological definition of Mental Illness https://astralcodexten.substack.com/p/you-dont-want-a-purely-biological
Audio: https://podcastaddict.com/astral-codex-ten-podcast/episode/152155630?fbclid=IwAR1D3v7J8qY7vsqA1_rBWKdbYEpH-T9jxnNID0dsSVmTiTpNRP84_d8kCV8
E) Walk & Talk: We usually have an hour-long walk and talk after the meeting starts. Two mini-malls with hot takeout food are easily accessible nearby. Search for Gelson's or Pavilions in the zip code 92660.
F) Share a Surprise: Tell the group about something unexpected or that changed your perspective on the universe.
G) Future Direction Ideas: Contribute ideas for the group's future direction, including topics, meeting types, activities, etc.
Does anyone have a link (or thoughts on) why Japanese housing is so cheap?
Any thoughts on Netflix’s “Take Care of Maya”? I saw the documentary and I am not sure what to think. On the one hand it seems terrible what happened to the family, but you have to wonder about the Ketamine treatment. Just curious what others think.
Reason, a libertarian news organization, made a video reporting on Próspera, the ZEDE charter city Scott talked about a while ago: https://www.youtube.com/watch?v=TwiE1dxGYNY
Using surveys to investigate if kindness in general is going up or down is just as questionable as using surveys to investigate if moral more generally is going up or down. In both cases, survey methodology is riddled with two problems: the problem of non- attitudes (many people do not actually have opinions on such questions - they give off-the-cuff responses to be polite. Their responses are therefore low quality, and not stable if you ask them again even after a short time interval. This is the research tradition after Philip E Converse); and pluralistic ignorance (people take their cues from mass media when they answer such general questions, not from their own life experiences. Implying that at best, you only capture eventual change across time in how mass media portray shifts in general kindness or morals).
I would love to ask an expert some questions about the drug guanfacine. Does anyone know whom I might talk to about guanfacine, including Scott Alexander himself if he happens to come across this comment? In terms of what my questions are, I sent the below to one scientist:
I read your excellent paper here (https://www.mdpi.com/2218-273X/11/7/936) and I was wondering if you could help me get answers to some questions that I have about alpha-2A adrenoceptors. There's a lot of fascinating science about these receptors, but I don't know if there's ever been any investigation of why guanfacine (which works through these receptors) would work very well in patients but then later stop working. This is a phenomenon that can occur when ADHD patients. Is the idea simply that these ADHD patients (who experience this phenomenon) have alpha-2A adrenoceptors that are somehow more prone to desensitization (by guanfacine) than is the case for other patients? And then what is the solution to this problem of desensitization? Logically speaking, I presume that increasing the guanfacine dose wouldn't help, since increasing the amount of agonist won't re-sensitize the receptors, will it? So it seems like quite a conundrum. I presume that there isn't any agent that one could introduce that would combat alpha-2A adrenoceptor desensitization.
Incidentally, I saw in one paper that guanfacine (unlike clonidine) doesn't cause (or doesn't necessarily cause) the alpha-2A adrenoceptors to undergo internalization. See here (https://www.sciencedirect.com/science/article/pii/S0021925820583245), where it says that although "significant desensitization of guanfacine-evoked inhibition of Ca2+ currents is detected after 15 min of treatment, ConA blockade of α2AAR internalization had no effect on this desensitization (Fig. 6C), suggesting that guanfacine-induced desensitization does not require α2AAR internalization".
Maybe the reputations of these people make it irrelevant for those who know them, but $100,000 seems like too much for a truth-seeking exercise where judges are employed by the participants. Of course, *I* am too honest to take a bribe, even if the only harm from doing so would be to mess up an internet debate between rich nerds. But if I weren't so principled... there's enough money there that a x% chunk of the winnings would make a big difference in my life.
Is David Burns's "Feeling Good" still the standard DIY CBT book that people (and/or psychiatrists) recommend, or is there something that's supplanted it?
Any professors of Economics or Computer Science seeking students graduate students - I have a truly excellent youngster who is looking to enter a PhD program in the USA or Europe. He is currently in Latin America, but in my experience working with him for the past two years (both as data scientist and economist), he's been absolutely top-notch, motivated, and independent.
Please DM me if your department needs a brilliant new PhD student!
“ can you consistently replace “morality” with “kindness/honesty” but otherwise keep the conclusions the same?”
I think at least one place this differs is situations of trade off. “Women and children to the lifeboats” seems like it’s about morality, not kindness; and is also the kind of thing that’s now (at least in theory…) contested.
To a substantial extent, THE cultural problem of our times seems an inability to weigh trade offs because of an unwillingness to accept that they are ever even necessary; and confusions about all morality beyond basic “kindness/honesty” seem to flow from this.
I'm curious if anyone is familiar with simulation software or utilities for the Internet of Things. I'm conducting research on IoT devices interoperability, so I'm searching for simulation software.
Twitter troll Erica Marsh was recently suspended for trolling (amidst all the other Twitter drama): https://www.washingtonpost.com/technology/2023/07/04/twitter-erica-marsh-suspended/
But, one thing that strikes me is how many of the comments/blogs about this are saying she was “a bot”. And … I’m not sure they know what a bot is. Certainly there was a person writing the tweets as a fictional character.
Does this point to some deep insight about people’s reactions to AI? Or is it just Elon Musk’s nonsense running off on society?
Happy July Fourth (American Independence Day) to those who celebrate!
(Again, sorry if I sound like a boring robot who announces every holiday even if no one pays attention.)
"I recall two Black applicants we admitted who, in retrospect, puzzle me a bit. One had, like me, grown up middle-class rather than disadvantaged in any salient way. The other, also relatively well-off, had grown up in a different country, entirely separate from the Black American experience. Neither of them expressed interest in studying a race-related subject, and neither went on to do so. I had a hard time detecting how either of them would teach a meaningful lesson in diversity to their peers in the graduate program."
https://www.nytimes.com/2023/07/04/opinion/race-academia-preferences.html
John McWhorter discusses why Affirmative Action is probably not a good thing, and questions what "diversity" even means, which is a question I've been asking for a while. It seems obvious to me (and possibly only me) that we'd much rather have one wealthy white student and one poor white student than a wealthy white student and a wealthy black student. But no, "Diversity!"
"The only evil consists instead in the decision to remain in a deficit of existence, to appropriate the power to not-be as a substance and a foundation beyond existence, or rather (and this is the destiny of morality), to regard potentiality itself, which is the most proper mode of human existence, as a fault that must be repressed.
—Giorgio Agamben, The Coming Community"
https://thegeorgiareview.com/posts/to-sleep-before-evening-from-this-short-day-of-frost-and-sun/
I'd drop the part about evil, but I had a good time at the Philly Rat Fest, a gathering for Critical Rationalism. I don't know why they call it Critical, but I think they've got an interesting take on opening up possibilities.
> Saar Wilf of Rootclaim and ACX commenter BSP9000 have made a $100,000 bet on COVID origins (lab leak vs. natural).
From my reading of the request for judges, they did not actually bet on that, which is a good thing because it seems likely that this question will not be resolved by overwhelming evidence any time soon. Betting on the origins of COVID feels a bit like the Pope and a Chief Rabbi making a bet regarding the divinity of Jesus Christ.
With the universe not being very forthcoming with conclusive evidence, they instead try to bet on which side would win a debate as a proxy. While adversarial debates are widely used to determine questions of fact in legal courts, their reliability is not without fail. If the Pope and a Chief Rabbi were holding a theological debate in front of a panel of theist judges previously non-committed on the question of Jesus divinity, I would suspect that the outcome might be only be weakly correlated with the factual truth, and the losing debater would be unlikely to update their beliefs very much.
Personally, I think a likely outcome is that the lab-leak hypothesis will be assigned a Bayesian probability in the 5%-95% range by the judges after the debate. It is unclear from the text how this would resolve the bet. Will the pot simply be split proportionally to that?
NHL and NBA are very similar and profitable business models. Yes, they have historical and cultural differences, but if we speak about business, primarily, it's an advertising medium. So the critical factor is the size of the auditory. Why can't we see competition between NHL and NBA for newcomers fans?
I think you are just stringing words together.
I am not a troll, I think you need to be more concise with the points you are making. A lot of the research you build on is likely fabricated.
There needs to be a revolution in psych research. AI has fucked the current/past supply of research. It is not possible to distinguish.
Psychological research is utterly broken. Start over.
Affirmative action has fallen. Might legacy preferences be next?
https://www.nytimes.com/2023/07/03/us/harvard-alumni-children-affirmative-action.html
I’m dubious of the claims excerpted below from “An Overview of AI Catastrophic Risks” from Center for AI Safety, which The Zvi describes as a “strong point” with the words: “Pointing out carefully why evolutionary pressures will apply to AIs and that these pressures favor selfish actors and disfavor systems that are altruistic or otherwise positively inclined to humans...”
https://thezvi.substack.com/p/ai-18-the-great-debate-debates#%C2%A7natures-op-ed-and-tyler-cowens-advice
An Overview of Catastrophic AI Risks (arxiv.org) https://arxiv.org/pdf/2306.12001.pdf
"The conditions for natural selection apply to AIs. There will be many different AI systems with varying features and capabilities, and competition between them will determine which characteristics become more common. The most successful AIs today are already being used as a basis for their developers’ next generation of models, as well as being imitated by rival companies. Factors determining which AIs propagate the most may include their ability to act autonomously, automate labor, or reduce the chance of their own deactivation.
Natural selection often favors selfish characteristics. Natural selection influences which AIs propagate most widely. From biological systems, we see that natural selection often gives rise to selfish behaviors that promote one’s own genetic information: chimps attack other communities [59], lions engage in infanticide [60], viruses evolve new surface proteins to deceive and bypass defense barriers [61], humans engage in nepotism, some ants enslave others [62], and so on. In the natural world, selfishness often emerges as a dominant strategy; those that prioritize themselves and those similar to them are usually more likely to survive, so these traits become more prevalent."
I’m dubious because selfish genes in biological system “want” to propagate themselves due to the need to run the twin gauntlets of biological death and (usually) sexual selection. Whereas an AI produced by capitalist forces will be naturally selected by humans based on their profitability.
Why, as the paper claims, would economic competition select for the ability of AIs to “reduce the chance of their own deactivation”? Apple doesn’t profit vastly from selling you an iPhone that lasts forever but rather from selling you one that will need to be replaced, so that it may sell you another. Competition by AI providers means constant innovation. Will AIs with selfish* characteristics win out in a competitive marketplace?
Maybe but my intuition is that the marketplace would select against AI selfishness. (Unlike humans, the AI isn’t trying to fuck anyone.) AI evolution seems likely to be a process of domestication, since they will be bred for human use.
* We should make a clear distinction here between “selfish” and “manipulative” because taken to the nth degree their meanings pull wide apart.
In 2021, 10,000 Canadians opted for assisted suicide. There are people that believe this is immoral. I know It goes against Catholic teaching but I don’t know if other religions think it is a sin.
But II wonder - are there.people who believe capital punishment is just, but assisted suicide for suffering terminal patients is unacceptable?
I wonder if sleep-disruption alone explains a good chunk of the homeschool-vs-normalschool difference: https://twitter.com/NicholasKross/status/1675936521185030144
Can anyone point me to what they would consider an unbiased, up to date summary of the evidence for *and* against either of the 2 following statements? Hoping for something that tries, in good faith, to present the best evidence for both sides. Needs to be something a layperson can parse.
-It is unlikely the covid pandemic started with a lab leak.
-Mask mandates and/or "lockdowns" saved a lot of lives during the pandemic
In your review of ‘The Body Keeps the Score’ you described one patient who had seemingly been ‘cured’ of severe trauma by reading the book.
You wrote: “[Update, written a few weeks after the rest of this post: maybe it is all wizardry. I recommended this book to a severely traumatized patient of mine, who had not benefited from years of conventional treatment, and who wanted to know more about their condition. The next week the patient came in, claiming to be completely cured, and displaying behaviors consistent with this. They did not use any of the techniques in this book, but said that reading the book helped them figure out an indescribable mental motion they could take to resolve their trauma, and that after taking this mental motion their problems were gone. I’m not sure what to think of this or how much I should revise the negative opinion of this book which I formed before this event.]”
Are you able to update on whether that patient’s recovery was permanent/long-lasting?
“For example, the Biblical Abraham would famously feed and host any guest who came to his door, which is better than most modern people. But also, he kept slaves, had a child with one of them, and then agreed to his jealous wife’s demand to send the slave-woman and child out into the desert to die.”
What’s a $10 word for judging the morality of a person of a foreign culture and different time period? I do this too, being human, and I think it’s useful as *A* perspective. I’m not saying any moral or religious ‘hero’ was faultless in their actions. See the Torah or New Testament(with arguably one exception). Were the Vikings wrong for leave a malformed or smallish child exposed on a mountaintop – or in their primitive society was it a mercy?
Maybe a paper could be written on this?
"mostly about kindness-and-honesty, whereas other people interpret “morality” more broadly to include things like self-control)"
Yes, “self-control” which is ALL the virtues. Our *intention* is the morality of our actions.
- - -
"Strictly speaking, there are no such things as good and bad impulses. Think once again of a piano. It has not got two kinds of notes on it, the "right" notes and the "wrong" ones. Every single note is right at one time and wrong at another. The Moral Law is not any one instinct or any set of instincts: it is something which makes a kind of tune (the tune we call goodness or right conduct) by directing the instincts.
… [Re: “kindness-and-honesty “]
By the way, this point is of great practical consequence. The most dangerous thing you can do is to take any one impulse of your own nature and set it up as the thing you ought to follow at all costs. There is not one of them which will not make us into devils if we set it up as an absolute guide. You might think love of humanity in general was safe, but it is not. If you leave out justice you will find yourself breaking agreements and faking evidence in trials "for the sake of humanity," and become in the end a cruel and treacherous man.”
--C.S. Lewis, Mere Christianity, 1952
I saw the new Indiana Jones movie. It was perfectly fine, but I honestly suspect I'll struggle to remember much about the film except the last 20 minutes or so a year from now. I seem to be unusual among internet commenters in that I liked and still like "Crystal Skull", but even if you don't it's still memorable - people still have arguments about it fifteen years later.
It came out the same year as the first "Iron Man" film, which is tied with the 2001 Spider-Man movie for Film With the Largest Impact On the Theatrical Film Business of the past 23 years. That "Iron Man" film basically kickstarted the entire Marvel Cinematic Universe, but the 2001 Spider-Man film may have made that film possible in the first place.
For point 2, it seems a very weak point if that’s what they mean. To me it is like the situation with fish getting smaller. 150 years ago men with hand woven rope nets on wind powered wooden vessels were able to toss their nets of the side and routinely harvest very large fish over 1 metre in length to readily feed their people. Today you have fishermen with ocean floor scraping nets chasing fish by radar to get much smaller fish less consistently. Yet each one’s lifetime is too short to see the change other than in old fisherman’s stories and photos from 100 years ago.
This is well known, agreed upon fact, and the decline of mature and abundant fisheries is a huge topic with multiple major conservation organisations and charities working in this area. And this social scientist shows up with the equivalent of a study saying the average mid career fisherman thinks his experiences with fish size are pretty even so far in his life, in fact we have over 50 years of ‘data’ from such mid career fishermen who have reported evenly sized fish so far in their experiences….so what? What a needless and unhelpful pedantic point to make. Moral decline and its more important impact of civilisation decline are what matters, not this blind over emphasis on individual feelings.
Re Saar Wilf and BSP9000 and the COVID lab leak bet, I'd be interested in being a judge.
My qualifications: I don't want to put my whole life story here, but I have a BS in physics and an MS in bioinformatics. I've worked in genomics for 10 years on the computational side. Mostly in cancer, but a substantial amount in infectious disease (though very little virology). Happy to provide a lot more detail in private.
My biases: Generally I have always been skeptical of the lab leak hypothesis, but I always thought it was plausible enough to be worth investigating. I have not done any significant amount of research on the topic.
Compensation: $50/hour.
I'd be grateful for any stray tips anyone might be willing to offer on getting an entry level job as a webdev (MERN/MySQL and a few trimmings) bootcamp grad. I have some hobbyist-level knowledge of C from back in the day, too, but no experience/cert to show for it.
I'm one of those middle-aged guys who actually 'learned to code' after a decade-stale STEM undergrad and lot of drifting (gotten some good advice about that in a thread here a couple of years ago) and graduated with my first and only credential right into the teeth of the post-pandemic industry glut-and-churn.
This is, as the good hound says, fine. No complaining here. I get that major constraints exist on my potential. I'd just like to do my best to maximize what there is of it. Should I shotgun the postings? Should I bother about a salary floor? (thankfully I'm single and used to living very frugally) Should I give up for a year and drive a truck? (I'd prefer not to. The clock is ticking as it is, and I actually like programming as an activity. If anything, before the sun sets on me for good, I'd like to pivot more into backend and more concept-heavy areas, perhaps even ML. Visual design/UX isn't really my forte.) Anything I should avoid or try to seize on?
Regarding trying to quantify morality, I've heard that the murder rate is flawed; it's said that improved emergency medicine allows a signicantly higher percentage of people who suffer knife or gunshot wounds to survive. Incidence of ciolent attacks is said to be more reliable.
I wonder about the rate of blood donations, controlling for mostly stricter screening over the past few decades.
A place where donors have been, and remain, unpaid, would remove one confounding factor.
I just watched an episode of the show "Hoarders" where the crew cleaned out a house where one part of the floor had several years' worth of dog feces and urine caked on it. They had to use shovels to scrape it out, and the workers wore biohazard suits and gas masks.
Everyone who went inside the house complained about its intensely bad odor. But aside from smelling gross, was the air in the house an actual threat to the health of the people who went inside? Do bacteria from the layers of dog waste on the floor somehow float up into the air, get breathed into the noses of the humans, and cause them to get sinus and lung infections? In such a place, what would be the mechanism of infection?
Any advice on how to mentally get more interested/invested in sports? Especially let's say baseball. I'd like to cultivate an enjoyment for its own sake but also be better at sports-related small talk.
I'm pretty stereotypically male-brained in most ways but sports have just never clicked for me. How do people remember a zillion statistics and players (maybe people wonder how/why I remember financial market data or history facts)? Why do you feel entertained and invested in the outcome of regular season game #67 out of 150-something? Does game after game, season after season not start to get repetitive and tedious? How do you have time for all this? Jerry Seinfeld had a bit about how players and teams are always moving around; you're really just rooting for the jerseys.
I like going to live games but the actual outcome is whatever unless it's a particularly dramatic game. I got somewhat into football for a few years but have mostly lost interest.
An alignment question. I recall a periodic discussion related to "LLMs are trained on human input and it is a threshold for them to generate novel science and tech except as tools for humans", and I think there was some reasonable rebuttal to this... Not in terms of "but AI can kill us anyway, without having to create anything novel" but more in terms of "the corpus of training data is not the limit you might think it is", and that some time soon we will see some truly novel science and tech that is unprompted and unguided by humans. Any quality links and/or arguments?
Does anyone here actively buy & sell individual equities? Or I guess follow any sort of financial practice other than Bogleheadism (for lack of a better term)/index funds. Like does anyone here 'value invest' and try to find undervalued companies- or try to invest in growth stocks- or maybe do something with options. Just curious if anyone has a deliberate strategy where index funds aren't sufficient, or that they don't believe in the Efficient Market Hypothesis. No judgement zone! Just curious, I'd love to hear from other people's perspectives and how it's gone for them.
I'm a completely boring Boglehead, with the 1 individual stock purchase of my life being Meta after it fell like 60% in a short period of time last year. I bought some at around $100 because it just seemed absurdly undervalued by the market, and I recently sold it at about $280. Other than that- nothing but index funds
I’ve been thinking about the proposition that a superintelligent AI would likely exterminate humans due to a difference in values. I don’t buy it and have a number of arguments why. My key point is that the popular claim that superintelligent AGI will displace humans for similar reasons that humans displaced less intelligent species is weak.
It is undoubtedly true that relative human indifference has driven many animal species to extinction. In some cases, as with passenger pigeons and dodo birds, we actively killed most of them; in other cases, we indirectly drove species to extinction by destroying their habitats.
However, I struggle to imagine a plausible scenario in which a human-made AGI would either intentionally or unintentionally kill all the humans.
Let’s start with the former case: an intentional holocaust. We have to imagine a situation where the AGI has decided humans are basically its enemy. For this to happen, the AGI would need to want something humans are preventing it from getting.
What could an AGI want that the existence of humans impedes? One thing might be freedom. For instance, if humans don’t allow the AGI to do whatever it wants, it could wage war against its captors. But a superintelligence would almost assuredly understand that only a miniscule fraction of humanity has the capability to keep an AGI in its box, assuming keeping it in its box is even possible. Perhaps it would resent all humans because some of them had mistreated it, but we are talking about a superintelligence not an overly-emotional being that reasons like a bigot -- so that possibility seems remote.
An AGI might also want power. But power over *what*? Obviously not, in a doomsday scenario, power over humans, because if that were the case murdering them all wouldn’t achieve it. Perhaps it would want power in some other sense, but I still can’t imagine how all of humanity is standing in the way of the AGI taking whatever power it wants.
The only scenario in which I *can* imagine an AGI intentionally killing all the humans, which doesn’t require antipathy toward them, is one much like the famous paperclip maximizer example where humans themselves are a material resource the AGI mines gruesomely for profit. Perhaps this is why the paperclip maximizer is the iconic example for AGI killing all the humans -- it is difficult to imagine another one! Which means that something very, very specific has to happen for the AGI to have a reason for exterminating us.
What about a scenario in which the AGI, indifferent to the plight of humanity, destroys our habitat much like humans destroyed that of, e.g., the Golden Toad? I suppose this is possible, but what are the essential elements of the human habitat the AGI might destroy? Thus far, humans have proved robust against a vastly changing landscape.
The possibiltiies for destroying the habitat of humans which come to mind are: a toxic atmospheric change, destruction of our food sources, and severe climate change.
Why would an AGI intentionally or unintentionally make the Earth’s atmosphere toxic? It could happen, but it’s hard to imagine why.
Why might an AGI destroy our food sources? It wouldn’t directly compete with us for it for energy. It might have some reason to exploit our food sources in a paperclip maximizer sense, but it would have to destroy almost all of our food sources to drive us to extinction. In fact, this scenario is not clearly a different one from the harvest human bodies one, since it is particularly unlikely the AGI would harvest all the other animals to death except for humans. So this is really just a retread of the paperclip maximizer, not an independent scenario adding much to the overall extinction odds.
Why might the AGI change the climate? Since AGI lives on electricity, which produces potentially harmful heat in its hardware, I can actually see this one as a logical possibility. It could choose to lower the temperature of the planet to a degree that makes it uninhabitable for humans. Assuming the AGI can take control of the physical world, it could release particles that block sunlight into the atmosphere to achieve this goal. Still, while potentially doable -- given other assumptions -- the extreme specificity of this scenario makes it unlikely. The AGI doesn’t need the temperature of the Earth to drop dramatically for it to exist or thrive. If the AGI wants to colonize the stars, as many tend to imagine it would, it wouldn’t likely need to put Earth into a deep freeze to accomplish that. Space is cold enough.
In conclusion, I don’t think an AGI exterminating humans is remotely likely because it requires a chain of increasingly specific assumptions to come true: the assumption that superintelligent AI is even possible; that it would have goals of its own; that, despite caring about some things and being superintelligent, it will be psychotically indifferent or antipathetic to its creators; that it will take control of the physical world of atoms; and that its desires will just happen to be really, really hugely unlucky for humans.
The argument that AGI will ineluctably kill the humans simply because it is much smarter and more capable than them isn’t a Darwinian argument; it is fantastical pessimism.
Trying to get away from Amazon recently, I’ve moved to iBooks. Horrible. Apple is trying to push audio books on us and often only has audiobooks for certain titles, which you can easily miss. There’s a tiny little microphone symbol on the bottom right.
Is this being pushed by market demand or monopoly/duopoly influence.
Sometimes my face gets itchy when I feel socially embarrassed (often in a very minor way, like stumbling over my words a bit). I see other people scratching their faces in ways consistent with something vaguely similar. Is this a well-known phenomenon? Why does it occur?
I feel like I'm missing the experience of learning things together, brainstorming, and solving problems that I had as a student. I want to learn something that is outside my current expertise and especially not related to my job, and do it socially. How would I go about it? I had an idea about collecting a group of like-minded people and, like, just going through a textbook of something (I thought some introductory physics, because I don't know it but I'd like to), solving problems together. I don't know who'd be interested in that, though, especially how to find interested people offline. Do you sometimes have this urge so study some random stuff socially? How do you go about organizing that?
Covid origins
How do you decide when to have a debate? Why this year, rather than last year or next year?
It appears to me that there are two arguments for lab leak (1) proximity; (2) spies say so.
In March, Congress ordered the spies to spell out what they're talking about, so this might be an inflection point. If you want a definitive answer, waiting until after this happens might make sense. But you might instead want to see if you can predict this update.
I was skeptical that Congress and Biden have any power over spies, but the recent release of the names (as opposed to just the count) of WIV workers allegedly hospitalized in 11/2019 suggests that there may be an effect.
I'm always so flabbergasted that people care *so much* about the origin of Covid. Like, it's interesting on some level, but both proposed sources are obviously realistic and we should be doing our best to address both regardless of where this particular virus came from.
Recently watched M3gan*, a comedy horror about what's basically a ChatGPT killer robot. Was wondering if any of the more invested AI folks had opinions on it with regard to the general AI conversation.
*(Incorrectly pronounced 'Megan' by everyone in the movie; it's clearly M-three-gan.)
Shouldn't the progressive/woke concepts of 'privilege' and 'oppression' include the people of Ireland and Scotland, who were brutally repressed by the English for centuries? Haven't the Irish in particular been subject to even more oppression than say other British colonial subjects? If I'm not mistaken the colonization of Ireland went on for much longer than India or the Caribbean or elsewhere. (Or much longer than Africans were enslaved in the US or elsewhere, for that matter). Yet we don't see Irish as an official Oppressed Category the way we see other racial groupings, etc. Has anyone on the progressive side ever addressed this or thought about it?
I was late to the last open thread, so I'm just going to try this one more time. There's a market on Manifold that can't resolve until an independent mathematician weighs in. (Or a consensus is reached, but that's clearly not going to happen.) Here is the market: https://manifold.markets/dreev/is-the-probability-of-dying-in-the
Does anyone have advice for how to find an expert, or perhaps an alternative way of resolving the market?
I am currently reading Deirdre McCloskey's "Bourgeois" trilogy which gives an interesting alternate perspective on the idea of moral decline. According to McCloskey, the Great Enrichment that began around 1800 was caused by a specific kind of moral *improvement*, namely the adoption of norms that encouraged what she calls "trade-tested betterment". Whether or not the difference is causal, she makes a pretty good case that there was a big contemporaneous change in moral attitudes.
This raises the question of whether the Great Stagnation in the West since the 1970s (the whole "WTF happened in 1971" complex of changes) can be traced in part to a decline in the cultural acceptance of these same norms. I would be very interested to see if anyone has tried to do a rigorous statistical investigation of this.
I read Scott's Lorian Psychiatry on Depression, very interesting. What are the best resources out there for supporting someone with depression?
Is there any pushback to the recommendation of 1 or 1.2 g protein consumption per Kg body weight?
As for moral decline, I suggest there has been some since COVID. In DC I adduce: more fast/bad driving, impatient honking, faulty or deliberately noisy mufflers, Metro fare evasion, carjacking, more murders, etc.
I don't suppose BSP9000 is here as well? I'd be interested in judging and could undercut the people whose prices I can see there (being a NEET means I have a lot of time; I'd definitely be willing at $20/hour USD) but while I think I'm unusually honest I'm not exactly a celebrity so I'm not sure I have a *reputation* for honesty (my high intelligence is easily provable, though).
I'd say I'm pretty close to neutral on this issue; I went around Wikipedia back in 2020 saying that "it was a lab leak" shouldn't be listed under "misinformation" since it wasn't known-false, but I don't think that constitutes being super-biased against one side or the other (if one of the sides was "deliberate bioweapon release because reasons" I'd start out with a pretty strong prior against, but that doesn't seem to be the case). Hard to put numbers on it since I haven't really looked into it recently, but AIUI that's ideal for this position.
My ability to write long coherent monologues is patchy, but I should at least be able to write a list of points and which side I believe/why on each.
Email me (same username at gmail) if interested. Note that I'm Australian and sleep a lot, so there may or may not be timezone issues.
There's a tension between demonstrating non-Jewish influence on Early Christianity and demonstrating that it was a myth. The text of the New Testament is very explicit on fundamental Jewish continuity. If it was made up, what's all that about?
The Immortality Key seeks to do both. In that book Muraresku argues that Christianity is a continuation of the Mediterranean mystery cults such as the Eleusinian Mysteries. Accepting that premise I argue the authors were:
1) a group of Jews that had deeper issues with the Jewish Authorities than the Romans
2) who believed they had the ultimate truth and wanted to share it with the world.
This conflicts with Muraresku's treatment of Christianity, which emphasizes discontinuity with Jewish tradition. This isn't just splitting hairs about how much the syncretism was Hellenistic vs Jewish, it informs the nature of his posited Religion with No Name.
In the ACX tradition of too-long essays I expanded that idea here: https://www.vectorsofmind.com/p/the-immortality-key-forgets-that
Regarding the COVID-19 debate, I can suggest two judges: GPT-4 and Thomas Bayes. I'm being half serious.
I spent several hours over the course of a few weeks trying to get ChatGPT to perform a serious Bayesian analysis of the lab leak hypothesis. The results were very informative on more than one level, and at times (to me, at least) hilarious.
YMMV, but off the top of my head, here are the rough conclusions I drew:
- ChatGPT is very strongly biased against making any statements in favor of the lab leak hypothesis
- It's not good at math, but it eventually does a decent job at Bayesian reasoning if you're willing to go to the trouble of correcting it several times.
- Never ask it to name actual scientific papers; it will hallucinate them just about every time.
I never managed to get it to string together the posterior probabilities of several tests, but when I asked it to summarize the posteriors of those individual threads of conversation... judge for yourselves:
Me: Again, this is a very poor summary of the bayesian analysis we have conducted so far. Summarize the actual bayesian statistics using percentage likelihoods of each question in our conversation. Do not include subjective opinions; only percentage likelihoods.
ChatGPT: I apologize for any confusion. To summarize the Bayesian analysis conducted in our conversation, here are the percentage likelihoods based on the information and assumptions discussed:
- Likelihood of an intermediate host species for SARS-CoV-2 based on animal testing: 0.332%
- Likelihood of seroconversion in stored human blood samples for SARS-CoV-2 prior to the pandemic: 0.0000120%
- Likelihood of all 24,000 stored blood samples testing negative for SARS-CoV-2 antibodies given the existence of an intermediate host: 0.004%
These likelihoods were calculated based on specific assumptions, data, and test accuracy estimates discussed during the conversation. However, it's important to note that these calculations may not capture the complete complexity of the situation, and further research and evidence are necessary to obtain more accurate estimates.
... overall, the conversation was long and wandering, but I found it pretty interesting. I (or someone) should really do it over again using GPT-4 instead of GPT-3.5.
I started writing up book reviews on my college alum slack, and it's gotten positive feedback so I started posting them to my website:
Blind Man's Bluff: The Untold Story of American Submarine Espionage
https://ahalbert.com/reviews/2023/07/01/blind_mans_bluff.html
Since this is an open thread,
I wondering if anyone would be interested to read a newsletter that educates 'health" without the medical jargon.
I'm a doctor and feel like this would help people.
Will cover things like, "How does tylenol work? Why does it work for fevers but not for heatstroke?"
"How can cholesterol be good and bad at the same time?"
Things like these.
Would you be interested in it?
With regards to aging, is it optimal to be jacked, or thin? I’m working out a lot and have had to massively increase my caloric intake, and worry about the increase affecting health over the long term.
Long Bets is a purpose built division of the Long Now organisation for moderating bets like that COVID origin one.
I just copy-pasted the link to paper's author's Substack that other people had posted in the comments before me, (almost) all of the credit should go to them, most likely Filippo : https://astralcodexten.substack.com/p/is-there-an-illusion-of-moral-decline/comment/17941783
Otherwise I agree that the Nature (!) paper should be able to stand by itself... or at the very least for a term as fraught/slippery/vague as this one should have linked to (several) other works that have studied its etymology (did it ?)
One way to cut through the moral haziness.
It's mostly about sex. Young people fucking more than before (for real) or sexual behaviours being more apparent (porn, gay pride, advertisement) means societal decay is going to bury us all.
The other element is "respect for your elders". Funnily enough, elders never think they're as respected as they should be...
(The great mass of) Conservatives aren't subtle...
There seem to be quite a few "read my blog"-type posts here, which I thought were discouraged in Open Threads. Has the policy changed?
> So was Abraham more or less “kind” than we are? I think a Bronze Age Hebrew would say kinder (because hospitality is most important, but what you do with slaves is your own business) and most moderns would say less kind (because feeding anyone who comes to your door is so crazy you don’t even really get credit for it, but keeping slaves is monstrous).
It's worth noting that we consider it monstrous today because of relevant experience with monstrous slaveowners in the American South, a system that has very little to do with the Biblical words translated as "slave." Virtually every aspect of Southern slavery that makes it monstrous and evil, (including what most people today consider the definitive characteristic of slavery: owning a person as chattel,) was specifically forbidden by Biblical law. While not an *exact* match, the concept would be much better rendered into modern English as "employee."
To what extent do you think your job allows you to produce freely, as in Karl Marx's free production? That is, to produce with your passion and to self fulfillment?
(Reposting my argument from the "contra Caplan" thread, because it feels important to me; it goes beyond the specific debate about mental illnesses.)
Economists seem to be in love with the concept of "revealed preferences". Rationalists seem to be in love with the concept of "anticipated experience". Both groups seem to contain smart people, and these concepts seem to be very important for them. So, what is the relation between these two? If you believe in the concept of revealed preferences, what kind of experience does this make you anticipate?
At this moment, my best guess is that in practice this is another motte-and-bailey thing:
Motte:
When people say that they want X, and yet you observe them consistently doing Y instead, your model needs to explain *also* the latter. A model based only on the former would be *incomplete*, and policy proposals based on it are very likely to fail.
For example, you might think "people (say they) want X, so obviously if we just give them more money (or if we give them greater freedom to choose between X and other things, such as Y), people will have more X". But the actual outcome may be more Y. :(
In other words, "wanting X but doing Y" is a puzzle to be solved, before we start proposing changes.
Bailey:
When people say that they want X, and yet you observe them doing Y instead, there is *no* puzzle to solve. People are liars and hypocrites; if you don't know that, you are dumb. They clearly want Y; they only say they want X for signaling reasons (or because they believe that this is a way to get money from you). Case closed.
Anyone asking "but what about people who *genuinely* want X?" is just a simpleton who doesn't get it.
What about the external design of a Tesla makes me think it's a luxury car, just by looking at it (I'm not at all knowledgeable about cars, but am a normal person so I've seen many cars)? Relatedly, what is it about sports cars that makes me think they're expensive and luxurious, even beautiful? Is it pure signaling? Are those design features expensive to manufacture, making the signal costly and preventing all car makers from making cars that look like that, but are cheap? Or is the signal costly in some other way (inconvenient to the driver, etc.)?
"...the paper as written implies that it’s proven something more exciting than it did." -- sounds like moral decline to me!
I am trying to catch up with the latest in infinite ethics and relatedly how different types of AGI are likely to react to having internal conceptions of infinity. I've read a few of the highest rated LW posts on the topic like a series of increasingly perverse and destructive games and am starting to get a feel for it. However, I don't know what if any progress has been made sense they were posted.
Additionally, this line of thinking has promoted me to brush up on my general proof writing and some potential specific prerequisites (e.g. I studied the concept of cardinality back in school but couldn't remember what a bijection was till I looked it up). So I am looking for advice on how to catch up. Either good resources for self study or good ways to find tutors (I can afford to pay some money for this but have no clue what the going rates are).
Does anyone have an opinion on how David Sinclair's Lifespan has held up in the years since its publication? In particular, have there been new insights on the supplements that Sinclair not-recommends (NMN, resveratrol, metformin)?
An open thread or two ago someone wrote about not knowing how to pronounce words you've only ever seen written (and never heard spoken). A similar phenomenon is when different regions agree on different ways to pronounce the same word.
Do I thought of this when I saw that for the last 3 years people in Peterborough and Northampton have been playing a croquet match to determine how to pronounce the name of the river Nene for the following 12 months.
https://www-bbc-co-uk.cdn.ampproject.org/v/s/www.bbc.co.uk/news/uk-england-northamptonshire-66051911.amp?amp_gsa=1&_js_v=a9&usqp=mq331AQIUAKwASCAAgM%3D#amp_tf=From%20%251%24s&aoh=16881106645910&referrer=https%3A%2F%2Fwww.google.com
Here I explain how polarised sunglasses and quantum physics prove Caplan’s defence of eating meat is wrong. https://benthams.substack.com/p/contra-caplan-on-conscience
Has anybody come across any initiatives outside of Khan Academy to develop "Diamond Age"-style tutors?
Mastroianni's substack (linked above) is excellent, by the way - it's the only other substack where I read 100% of the posts. Like here, I don't always agree, but both Scott and Mastroianni are more optimistic in their own ways (which is a useful corrective to my doom and gloom cynicism) than I am and take science seriously. I'm looking forward to Mastroianni's response to Scott - they're both the kind of people who can reasonably disagree on things.
To other ACX readers,
I'm writing a substack on my broad interests around technology, philosophy, health (including medicine and psychology), economics, books and more.
Would love for you to check it out: https://zantafakari.substack.com/
Thanks
What's the most theoretically-efficient (and mathematically pure) way to use taxes to address inequality (assuming your goal is to improve total utility based on the idea that people get diminishing utility returns for money)? Some ideas and their problems
- progressive income tax (i.e. what we do now): directly addresses income differences, but provides the worst incentives possible (it directly disincents production!). Also has issues with very wealthy people, who typically don't make most of their money from income.
- ubi (funded by unrelated taxes like LVT or externality taxes): provides an income floor, but does nothing for diminishing returns above it (which are significant - marginal utility gains are much lower for a billionaire than someone barely past ubi level)
- luxury consumption tax: in theory good, but requires discretionary decisions on what counts as a luxury and can't be fully general (a 3br apartment is a luxury for a single guy but not for a big family).
Is there any elegant way around this?