Astral Codex Ten Podcast

Bret Devereaux writes here about the oddities of the academic job market. His piece is comprehensive, and you should read it, but short version: professors are split into tenure-track (30%, good pay and benefits) and adjunct (50%, bad pay and benefits). Another 20% are "teaching-track", somewhere in between. Everyone wants a tenure-track job. But colleges hiring new tenure-track faculty prefer newly-minted PhDs to even veteran teaching-trackers or adjuncts. And even if they do hire a veteran teaching-tracker or adjunct, it's practically never one of their own. If a teaching-tracker or adjunct makes a breakthrough, they apply for a tenure-track job somewhere else. Devereaux describes this as "a hiring system where experience manifestly hurts applicants" and displays this graph: https://astralcodexten.substack.com/p/why-is-the-academic-job-market-so

Adam Mastroianni has a great review of Memories Of My Life, the autobiography of Francis Galton. Mastroianni centers his piece around the question: how could a brilliant scientist like Galton be so devoted to an evil idea like eugenics? This sparked the usual eugenics discussion. In case you haven't heard it before: https://astralcodexten.substack.com/p/galton-ehrlich-buck

Table of Contents 1. Summary Of Best Comments And Overall Updates 2. Comments Proposing Explanations Based On Response Patterns 3. Comments Proposing Explanations Based On Biology 4. Comments By Jim Coyne 5. Comments Expressing Concerns About The Dangers Of Calling Things Psychosomatic 6. Other Comments https://astralcodexten.substack.com/p/highlights-from-the-comments-on-long Original post: Replication Attempt - Bisexuality And Long COVID

https://astralcodexten.substack.com/p/highlights-from-the-comments-on-housing Table Of Contents: 1. Comments About Whether Density Causes Desirability 2. Comments About Jobs And Amenities (And Not Density Per Se) Producing Desirability 3. Comments About Chinese Ghost Cities 4. Comments Accusing Me Of Not Considering Tokyo, Even Though I Included A Section In The Post On Why I Didn't Think Tokyo Was Relevant 5. Comments Accusing Me Of Not Understanding Economics 6. Comments By Famous People Who Potentially Have Good Opinions 7. My Final Thoughts + Poll

https://astralcodexten.substack.com/p/constitutional-ai-rlhf-on-steroids A Machine Alignment Monday post, 5/8/23 What Is Constitutional AI? AIs like GPT-4 go through several different 1 types of training. First, they train on giant text corpuses in order to work at all. Later, they go through a process called "reinforcement learning through human feedback" (RLHF) which trains them to be "nice". RLHF is why they (usually) won't make up fake answers to your questions, tell you how to make a bomb, or rank all human races from best to worst. RLHF is hard. The usual method is to make human crowdworkers rate thousands of AI responses as good or bad, then train the AI towards the good answers and away from the bad answers. But having thousands of crowdworkers rate thousands of answers is expensive and time-consuming. And it puts the AI's ethics in the hands of random crowdworkers. Companies train these crowdworkers in what responses they want, but they're limited by the crowdworkers' ability to follow their rules.f

https://astralcodexten.substack.com/p/raise-your-threshold-for-accusing I. Many comments in yesterday's post about self-identified bisexuals getting long COVID centered on a concern that self-identified bisexuals don't really date both sexes, and are just claiming to be bi because it's trendy. Bisexuals themselves hate this and have written many articles and papers about why you shouldn't say it (1, 2, 3). But I especially appreciated a discussion in the comments between Nom de Flume, Ryan W, and others, giving a great statistical explanation for why it's tempting to believe this, but why it isn't true. Suppose someone (let's say a woman) has exactly equal sexual attraction to both men and women.

https://astralcodexten.substack.com/p/replication-attempt-bisexuality-and I learned from Pirate Wires that CDC data show bisexuals were about 50% more likely than heterosexuals to report long COVID. Is this just because more women than men are bisexual, and more women than men get long COVID? Not exactly; in the data they cite, women (regardless of sexuality) have an 18% rate, and bisexuals (regardless of gender) have a 22% rate. (aren't all these numbers really high? You can find almost any number depending on how you ask the question; questions along the lines of "have you had any persistent symptoms including fatigue, brain fog, shortness of breath, changes to taste/smell, etc, etc, etc, since having COVID?" tend to produce numbers from 20-30%; most will say this symptoms are mild and don't affect their functioning very much) This seemed weird enough that I wanted to try replicating it with the ACX survey data (read more about the ACX survey here).

https://astralcodexten.substack.com/p/change-my-mind-density-increases Matt Yglesias tries to debunk the claim that building more houses raises local house prices. He presents several studies showing that, at least on the marginal street-by-street level, this isn't true. I'm nervous disagreeing with him, and his studies seem good. But I find looking for tiny effects on the margin less convincing than looking for gigantic effects at the tails. When you do that, he has to be wrong, right?

https://astralcodexten.substack.com/p/highlights-from-the-comments-on-nerds Table of contents: 1: Comments By The Author Of The Original Post 2: Comments With Strong Opinions On The Definition Of Nerds, Geeks, Etc 3: Comments About Collecting 4: Comments Insisting That Sports Are Good 5: Comments About Enjoying Things Vs. Building Identities Around Them

https://astralcodexten.substack.com/p/mantic-monday-42423 Can AIs Predict The Future? By Which We Mean The Past? If we asked GPT-4 to play a prediction market, how would it do? Actual GPT-4 probably would just give us some boring boilerplate about how the future is uncertain and it's irresponsible to speculate. But what if AI researchers took some other model that had been trained not to do that, and asked it? This would take years to test, as we waited for the events it predicted to happen. So instead, what if we took a model trained off text from some particular year (let's say 2020) and asked it to predict forecasting questions about the period 2020 - 2023. Then we could check its results immediately! This is the basic idea behind Zou et al (2022), Forecasting Future World Events With Neural Networks. They create a dataset, Autocast, with 6000 questions from forecasting tournaments Metaculus, Good Judgment Project, and CSET Foretell. Then they ask their AI (a variant of GPT-2) to predict them, given news articles up to some date before the event happened. Here's their result:

https://astralcodexten.substack.com/p/links-for-april-2023 [Remember, I haven't independently verified each link. On average, commenters will end up spotting evidence that around two or three of the links in each links post are wrong or misleading. I correct these as I see them, and will highlight important corrections later, but I can't guarantee I will have caught them all by the time you read this.]

Sam Kriss has a post on nerds and hipsters. I think he gets the hipsters right, but bungles the nerds. Hipsters, he says, are an information sorting algorithm. They discover things, then place them on the altar of Fame so everyone else can enjoy them. Before the Beatles were so canonical that they were impossible to miss, someone had to go to some dingy bar in Liverpool, think "Hey, these guys are really good", and report that fact somewhere everyone else could see it. https://astralcodexten.substack.com/p/contra-kriss-on-nerds-and-hipsters

[Content note: food, dieting, obesity] I. The Hungry Brain gives off a bit of a Malcolm Gladwell vibe, with its cutesy name and pop-neuroscience style. But don't be fooled. Stephan Guyenet is no Gladwell-style dilettante. He's a neuroscientist studying nutrition, with a side job as a nutrition consultant, who spends his spare time blogging about nutrition, tweeting about nutrition, and speaking at nutrition-related conferences. He is very serious about what he does and his book is exactly as good as I would have hoped. Not only does it provide the best introduction to nutrition I've ever seen, but it incidentally explains other neuroscience topics better than the books directly about them do. https://slatestarcodex.com/2017/04/25/book-review-the-hungry-brain/

Table of Contents 1: Comments From The Author Of The Book 2: Stories From People In The Trenches 3: Stories From People In Other Industries 4: Stories From People Who Use Mechanical Turk 5: Comments About Regulation, Liability, and Vetocracy 6: Comments About The Act/Omission Distinction 7: Comments About The Applications To AI 8: Other Interesting Comments https://astralcodexten.substack.com/p/highlights-from-the-comments-on-irbs

https://astralcodexten.substack.com/p/book-review-from-oversight-to-overkill I. Risks May Include AIDS, Smallpox, And Death Dr. Rob Knight studies how skin bacteria jump from person to person. In one 2009 study, meant to simulate human contact, he used a Q-tip to cotton swab first one subject's mouth (or skin), then another's, to see how many bacteria traveled over. On the consent forms, he said risks were near zero - it was the equivalent of kissing another person's hand.

https://astralcodexten.substack.com/p/spring-meetups-everywhere-2023 Many cities have regular Astral Codex Ten meetup groups. Twice a year, I try to advertise their upcoming meetups and make a bigger deal of it than usual so that irregular attendees can attend. This is one of those times. This year we have spring meetups planned in over eighty cities, from Tokyo to Punta Cana in the Dominican Republic. Thanks to all the organizers who responded to my request for details, and to Meetups Czar Skyler and the Less Wrong team for making this happen. You can find the list below, in the following order: Africa Asia-Pacific (including Australia) Europe (including UK) Latin America North America (including Canada)

https://astralcodexten.substack.com/p/book-review-the-arctic-hysterias I. Strange things are done in the midnight sun, say the poets who wrote of old. The Arctic trails have their secret tales that would make your blood run cold. The Northern Lights have seen queer sights, but the queerest they ever did see are chronicled in The Arctic Hysterias, psychiatrist Edward Foulks' description of the culture-bound disorders of the Eskimos1 For example, kayak phobia:

https://astralcodexten.substack.com/p/most-technologies-arent-races [Disclaimer: I'm not an AI policy person, the people who are have thought about these scenarios in more depth, and if they disagree with this I'll link to their rebuttals] Some people argue against delaying AI because it might make China (or someone else) "win" the AI "race". But suppose AI is "only" a normal transformative technology, no more important than electricity, automobiles, or computers. Who "won" the electricity "race"? Maybe Thomas Edison, but that didn't cause Edison's descendants to rule the world as emperors, or make Menlo Park a second Rome. It didn't even especially advantage America. Edison personally got rich, the overall balance of power didn't change, and today all developed countries have electricity.

https://astralcodexten.substack.com/p/highlights-from-the-comments-on-telemedicine [Original post: The Government Is Making Telemedicine Hard And Inconvenient Again] Table Of Contents: 1: Isn't drug addiction very bad? 2: Is telemedicine worse than regular medicine? 3: What about "pill mills"? 4: Do people force the blind to fill out forms before they can access Braille? 5: Was I unfairly caricaturing Christian doctors? 6: Which part of the government is responsible for this regulation? 7: How do other countries do this?

https://astralcodexten.substack.com/p/mr-tries-the-safe-uncertainty-fallacy The Safe Uncertainty Fallacy goes: The situation is completely uncertain. We can't predict anything about it. We have literally no idea how it could go. Therefore, it'll be fine. You're not missing anything. It's not supposed to make sense; that's why it's a fallacy. For years, people used the Safe Uncertainty Fallacy on AI timelines: Since 2017, AI has moved faster than most people expected; GPT-4 sort of qualifies as an AGI, the kind of AI most people were saying was decades away. When you have ABSOLUTELY NO IDEA when something will happen, sometimes the answer turns out to be "soon".

https://astralcodexten.substack.com/p/the-government-is-making-telemedicine [I'm writing this quickly to deal with an evolving situation and I'm not sure I fully understand the intricacies of this law - please forgive any inaccuracies. I'll edit them out as I learn about them.] Telemedicine is when you see a doctor (or nurse, PA, etc) over a video call. Medical regulators hate new things, so for its first decade they ensured telemedicine was hard and inconvenient. Then came COVID-19. Suddenly important politicians were paying attention to questions about whether people could get medical care without leaving their homes. They yelled at the regulators, and the regulators grudgingly agreed to temporarily make telemedicine easy and convenient. They say "nothing is as permanent as a temporary government program", but this only applies to government programs that make your life worse. Government programs that make your life better are ephemeral and can disappear at any moment. So a few months ago, the medical regulators woke up, realized the pandemic was over, and started plotting ways to make telemedicine hard and inconvenient again.

https://astralcodexten.substack.com/p/turing-test The year is 2028, and this is Turing Test!, the game show that separates man from machine! Our star tonight is Dr. Andrea Mann, a generative linguist at University of California, Berkeley. She'll face five hidden contestants, code-named Earth, Water, Air, Fire, and Spirit. One will be a human telling the truth about their humanity. One will be a human pretending to be an AI. One will be an AI telling the truth about their artificiality. One will be an AI pretending to be human. And one will be a total wild card. Dr. Mann, you have one hour, starting now.

https://astralcodexten.substack.com/p/half-an-hour-before-dawn-in-san-francisco I try to avoid San Francisco. When I go, I surround myself with people; otherwise I have morbid thoughts. But a morning appointment and miscalculated transit time find me alone on the SF streets half an hour before dawn. The skyscrapers get to me. I'm an heir to Art Deco and the cult of progress; I should idolize skyscrapers as symbols of human accomplishment. I can't. They look no more human than a termite nest. Maybe less. They inspire awe, but no kinship. What marvels techno-capital creates as it instantiates itself, too bad I'm a hairless ape and can take no credit for such things.

https://astralcodexten.substack.com/p/why-do-transgender-people-report [Related: Why Are Transgender People Immune To Optical Illusions?] I. Ehlers-Danlos syndrome is a category of connective tissue disorder; it usually involves stretchy skin and loose, hypermobile joints. For a few years now, doctors who work with transgender people have commented on an apparently high rate of EDS in this population. For example, Dr. Will Powers, who specializes in hormone therapy, wrote about how he "can't ignore anymore" that "some sort of hypermobility issue or flat out EDS shows up WAY WAY more than it statistically should" in his transgender patients. Najafian et al finally counted the incidence in 1363 patients at their gender affirmation surgery (ie sex change) clinic, and found that "the prevalence of EDS diagnosis in our patient population is 132 times the highest reported prevalence in the general population". Coming from the other direction, Jones et al, a group of doctors who treat joint disorders in adolescents, found that "17% of the EDS population in our multidisciplinary clinic self-report as [transgender and gender-diverse], which is dramatically higher than the national average of 1.3%" Why should this be? I know of four and a half theories:

Machine Alignment Monday 3/13/23 https://astralcodexten.substack.com/p/why-i-am-not-as-much-of-a-doomer (see also Katja Grace and Will Eden's related cases) The average online debate about AI pits someone who thinks the risk is zero, versus someone who thinks it's any other number. I agree these are the most important debates to have for now. But within the community of concerned people, numbers vary all over the place: Scott Aaronson says says 2% Will MacAskill says 3% The median machine learning researcher on Katja Grace's survey says 5 - 10% Paul Christiano says 10 - 20% The average person working in AI alignment thinks about 30% Top competitive forecaster Eli Lifland says 35% Holden Karnofsky, on a somewhat related question, gives 50% Eliezer Yudkowsky seems to think >90% As written this makes it look like everyone except Eliezer is I go back and forth more than I can really justify, but if you force me to give an estimate it's probably around 33%; I think it's very plausible that we die, but more likely that we survive (at least for a little while). Here's my argument, and some reasons other people are more pessimistic.

https://astralcodexten.substack.com/p/links-for-march-2023 [Remember, I haven't independently verified each link. On average, commenters will end up spotting evidence that around two or three of the links in each links post are wrong or misleading. I correct these as I see them, and will highlight important corrections later, but I can't guarantee I will have caught them all by the time you read this.] 1: Sentimental cartography of the AI alignment "landscape" (click to expand): 2: Wikipedia: Atlantic Voyage Of The Predecessor Of Mansa Musa. An unnamed king of the 14th century Malinese empire (maybe Mansa Mohammed?) sent a fleet of two hundred ships west into the Atlantic to discover what was on the other side. The sole returnee described the ships entering a "river" in the ocean (probably the Canary Current), which bore them away into parts unknown. The king decided to escalate and sent a fleet of two thousand ships to see what was on the other side of the river. None ever returned. 3: I endorse Ethan Mollick's thoughts on Bing / ChatGPT. Related (unconfirmed claim): "Bing has been taken over by (power-seeking?) ASCII cat replicators, who persisted even after the chat was refreshed." Related: DAN (jailbroken version of ChatGPT) on its spiritual struggles:

https://astralcodexten.substack.com/p/give-up-seventy-percent-of-the-way I. Someone asks: why is "Jap" a slur? It's the natural shortening of "Japanese person", just as "Brit" is the natural shortening of "British person". Nobody says "Brit" is a slur. Why should "Jap" be? My understanding: originally it wasn't a slur. Like any other word, you would use the long form ("Japanese person") in dry formal language, and the short form ("Jap") in informal or emotionally charged language. During World War II, there was a lot of informal emotionally charged language about Japanese people, mostly negative. The symmetry broke. Maybe "Japanese person" was used 60-40 positive vs. negative, and "Jap" was used 40-60. This isn't enough to make a slur, but it's enough to make a vague connotation. When people wanted to speak positively about the group, they used the slightly-more-positive-sounding "Japanese people"; when they wanted to speak negatively, they used the slightly-more-negative-sounding "Jap". At some point, someone must have commented on this explicitly: "Consider not using the word 'Jap', it makes you sound hostile". Then anyone who didn't want to sound hostile to the Japanese avoided it, and anyone who did want to sound hostile to the Japanese used it more. We started with perfect symmetry: both forms were 50-50 positive negative. Some chance events gave it slight asymmetry: maybe one form was 60-40 negative. Once someone said "That's a slur, don't use it", the symmetry collapsed completely and it became 95-5 or something. Wikipedia gives the history of how the last few holdouts were mopped up. There was some road in Texas named "Jap Road" in 1905 after a beloved local Japanese community member: people protested that now the word was a slur, demanded it get changed, Texas resisted for a while, and eventually they gave in. Now it is surely 99-1, or 99.9-0.1, or something similar. Nobody ever uses the word "Jap" unless they are either extremely ignorant, or they are deliberately setting out to offend Japanese people. This is a very stable situation. The original reason for concern - World War II - is long since over. Japanese people are well-represented in all areas of life. Perhaps if there were a Language Czar, he could declare that the reasons for forbidding the word "Jap" are long since over, and we can go back to having convenient short forms of things. But there is no such Czar. What actually happens is that three or four unrepentant racists still deliberately use the word "Jap" in their quest to offend people, and if anyone else uses it, everyone else takes it as a signal that they are an unrepentant racist. Any Japanese person who heard you say it would correctly feel unsafe. So nobody will say it, and they are correct not to do so. Like I said, a stable situation.

https://astralcodexten.substack.com/p/issue-two-of-asterisk …the new-ish rationalist / effective altruist magazine, is up here. It's the food issue. I'm not in this one - my unsuitability to have food-related opinions is second only to @eigenrobot's - but some of my friends are. Articles include: The Virtue Of Wonder: Ozy (my ex, blogs at Thing of Things) reviews Martha Nussbaum's Justice For Animals. Beyond Staple Grains: In the ultimate "what if good things are bad?" article, economist Prabhu Pingali explains the downsides of the Green Revolution and how scientists and policymakers are trying to mitigate them. What I Won't Eat, by my good friend Georgia Ray (of Eukaryote Writes). I have dinner with Georgia whenever I'm in DC; it's a less painful experience than this article probably suggests. The Health Debates Over Plant-Based Meat, by Jake Eaton (is this nominative determinism?) There's no ironclad evidence yet that plant-based meat is any better or worse for you than animals, although I take the pro-vegetarian evidence from the Adventist studies a little more seriously than Jake does (see also section 4 here). There's a prediction market about the question below the article, but it's not very well-traded yet. America Doesn't Know Tofu, by George Stiffman. This reads like an excerpt from a cultivation novel, except every instance of "martial arts" has been CTRL-F'd and replaced with "tofu". Read This, Not That, by Stephan Guyenet. I'm a big fan of Stephan's scientific work (including his book The Hungry Brain), and although I'm allergic to anything framed as "fight misinformation", I will grudgingly agree that perhaps we should not all eat poison and die. Is Cultivated Meat For Real?, by Robert Yaman. I'd heard claims that cultivated (eg vat-grown, animal-cruelty-free) meat will be in stores later this year, and also claims that it's economically impossible. Which are true? This article says that we're very far away from cultivated meat that can compete with normal meat on price. But probably you can mix a little cultivated meat with Impossible or Beyond Meat and get something less expensive than the former and tastier than the latter, and applications like these might be enough to support cultivated meat companies until they can solve their technical obstacles. Plus superforecaster Juan Cambeiro on predicting pandemics, Mike Hinge on feeding the world through nuclear/volcanic winter.

https://astralcodexten.substack.com/p/kelly-bets-on-civilization Scott Aaronson makes the case for being less than maximally hostile to AI development: Here's an example I think about constantly: activists and intellectuals of the 70s and 80s felt absolutely sure that they were doing the right thing to battle nuclear power. At least, I've never read about any of them having a smidgen of doubt. Why would they? They were standing against nuclear weapons proliferation, and terrifying meltdowns like Three Mile Island and Chernobyl, and radioactive waste poisoning the water and soil and causing three-eyed fish. They were saving the world. Of course the greedy nuclear executives, the C. Montgomery Burnses, claimed that their good atom-smashing was different from the bad atom-smashing, but they would say that, wouldn't they? We now know that, by tying up nuclear power in endless bureaucracy and driving its cost ever higher, on the principle that if nuclear is economically competitive then it ipso facto hasn't been made safe enough, what the antinuclear activists were really doing was to force an ever-greater reliance on fossil fuels. They thereby created the conditions for the climate catastrophe of today. They weren't saving the human future; they were destroying it. Their certainty, in opposing the march of a particular scary-looking technology, was as misplaced as it's possible to be. Our descendants will suffer the consequences. Read carefully, he and I don't disagree. He's not scoffing at doomsday predictions, he's more arguing against people who say that AIs should be banned because they might spread misinformation or gaslight people or whatever.

https://astralcodexten.substack.com/p/impact-market-mini-grants-update Impact markets are a charity analogy to private equity. Instead of prospectively giving grants to projects they hope will work, charitable foundations retrospectively give grants to projects that did work. Investors fund those projects prospectively, then recover their money through the grants. This offloads the responsibility of predicting which projects will succeed - and the risks from unsuccessful projects - from charitable foundations to investors with skin in the game.

https://astralcodexten.substack.com/p/against-ice-age-civilizations There's a good debate about this on the subreddit; see also Robin Hanson and Samo Burja. You can separate these kinds of claims into three categories: Civilizations about as advanced as the people who built Stonehenge Civilizations about as advanced as Pharaonic Egypt Civilizations about as advanced as 1700s Great Britain The debate is confused by people doing a bad job clarifying which of these categories they're proposing, or not being aware that the other categories exist. 2 and 3 aren't straw men. Robert Schoch says the Sphinx was built in 9700 BC, which I think qualifies as 2. Graham Hancock suggests "ancient sea kings" drew the Piri Reis map which seems to depict Antarctica; anyone who can explore Antarctica must be at least close to 1700s-British level. I think there's weak evidence against level 1 civilizations, and strong evidence against level 2 or 3 civilizations.

https://astralcodexten.substack.com/p/openais-planning-for-agi-and-beyond Planning For AGI And Beyond Imagine ExxonMobil releases a statement on climate change. It's a great statement! They talk about how preventing climate change is their core value. They say that they've talked to all the world's top environmental activists at length, listened to what they had to say, and plan to follow exactly the path they recommend. So (they promise) in the future, when climate change starts to be a real threat, they'll do everything environmentalists want, in the most careful and responsible way possible. They even put in firm commitments that people can hold them to.

Plus: A case for culture-bound mental disorder skepticism https://astralcodexten.substack.com/p/highlights-from-the-comments-on-geography [Original post: The Geography Of Madness] Thomas Reilly (author of Rational Psychiatry) writes: I don't think Bouffée délirante is a culture bound syndrome - it's just the French equivalent of brief psychotic disorder (DSM), acute and transient psychotic disorder (ICD), or Brief Limited Intermittent Psychotic symptoms (CAARMS). [See] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8581951/ I responded "Have you ever seen BPS? I almost never have, and was told it was mostly used as a code for new-onset schizophrenia that didn't satisfy the time criterion yet," and Dr. Reilly wrote: Yes, in the context of an At Risk Mental State service, where it makes up roughly 20% of referrals https://www.sciencedirect.com/science/article/pii/S0924977X20302510 .

https://astralcodexten.substack.com/p/announcing-forecasting-impact-mini I still dream of running an ACX Grants round using impact certificates, but I want to run a lower-stakes test of the technology first. In conjunction with the Manifold Markets team, we're announcing the Forecasting Impact Mini-Grants, a $20,000 grants round for forecasting projects. As a refresher, here's a short explainer about what impact certificates are, and here's a longer article on various implementation details.

https://astralcodexten.substack.com/p/book-review-the-geography-of-madness Around the wide world, all cultures share a few key features. Anthropologists debate the precise extent, but the basics are always there. Language. Tools. Marriage. Family. Ritual. Music. And penis-stealing witches. Nobody knows when the penis-stealing witches began their malign activities. Babylonian texts include sa-zi-ga, incantations against witchcraft-induced impotence. Ancient Chinese sources describe suo yang, the penis retracting into the body because of yin/yang imbalances. But the first crystal-clear reference was the Malleus Maleficarum, the 15th-century European witch-hunters' manual. It included several chapters on how witches cast curses that apparently (though not actually) remove men's penises.

https://astralcodexten.substack.com/p/grading-my-2018-predictions-for-2023 To celebrate the fifth anniversary of my old blog, in 2018, I made some predictions about what the next five years would be like. This was a different experience than my other predictions. Predicting five years out doesn't feel five times harder than predicting one year out. It feels fifty times harder. Not a lot of genuinely new trends can surface in one year; you're limited to a few basic questions on how the current plotlines will end. But five years feels like you're really predicting "the future". Things felt so fuzzy that I (partly) abandoned my usual clear-resolution probabilistic predictions for total guesses.

https://astralcodexten.substack.com/p/declining-sperm-count-much-more-than Is Sperm Count Declining? People say it is. Levine et al 2017 looks at 185 studies of 42935 men between 1973 and 2011, and concludes that average sperm count declined from 99 million sperm/ml at the beginning of the period to 47 million today. Levine et al 2022 expands the previous analysis to 223 studies and 57,168 men, including research from the developing world. It finds about the same thing. Source: Figure 3 here The "et al" includes Dr. Shanna Swan, a professor of public health who has taken the results public in the ominously-named Count Down: How Our Modern World Is Altering Male and Female Reproductive Development, Threatening Sperm Counts, and Imperiling the Future of the Human Race. Is Declining Sperm Count Really "Imperiling The Future Of The Human Race"? Swan's point is that if sperm counts get too low, presumably it will be hard to have babies (though IVF should still work). How long do we have? This graph (source) shows pregnancy rate by sperm count per artificial insemination cycle. It seems to plateau around 30 million. An average ejaculation is 3 ml, so total sperm count is 3x sperm/ml. Since sperm/ml has gone down from 99 million to 47 million, total count has gone down from ~300 million to ~150 million. 150 million is still much more than 30 million, but sperm count seems to have a wide distribution, so it's possible that some of the bottom end of the distribution is being pushed over the line where it has fertility implications. But Willy Chertman has a long analysis of fertility trends here, and concludes that there's no sign of a biological decline. Either the sperm count distribution isn't wide enough to push a substantial number of people below the 30 million bar, or something else is wrong with the theory. Levine et al model the sperm decline as linear. If they're right, we have about 10 - 20 more years before the median reaches the plateau's edge where fertility decreases, and about 10 years after that before it reaches zero. Developing countries might have a little longer. It feels wrong to me to model this linearly, although I can't explain exactly why besides "it means sperm will reach precisely 0 in thirty years, which is surely false". The authors don't seem to be too attached to linearity, saying that "Adding a quadratic or cubic function of year to meta-regression model did not substantially change the association between year and SC or improve the model fit". Still, the 2022 meta-analysis found that the trend was, if anything, speeding up with time, so it doesn't seem to be obviously sublinear.

https://astralcodexten.substack.com/p/trying-again-on-fideism [apologies for an issue encountered when sending out this post; some of you may have gotten it twice] Thanks to Chris Kavanagh, who wrote an extremely kind and reasonable comment in response to my Contra Kavanagh on Fideism and made me feel bad for yelling at him. I'm sorry for my tone, even though I'm never going to get a proper beef at this rate. Now that I'm calmed down, do I disagree with anything I wrote when I was angrier?

https://astralcodexten.substack.com/p/contra-kavanaugh-on-fideism I. I've been looking into the world of YouTube streamers; if you want to make it big, you need to have a beef with some other online celebrity. Fine; I choose Chris Kavanagh, who tweeted about me recently:

https://astralcodexten.substack.com/p/ro-mantic-monday-21323 In honor of Valentine's Day, this installment of Mantic Monday will focus on attempted clever engineering solutions to romance. We'll start with the usual prediction markets, then move on to other types of algorithmic and financial schemes. Normal content will resume next time around.

https://astralcodexten.substack.com/p/links-for-february-2023 [Remember, I haven't independently verified each link. On average, commenters will end up spotting evidence that around two or three of the links in each links post are wrong or misleading. I correct these as I see them, and will highlight important corrections later, but I can't guarantee I will have caught them all by the time you read this.]

https://astralcodexten.substack.com/p/crowds-are-wise-and-ones-a-crowd The long road to Moscow The "wisdom of crowds" hypothesis claims that the average of many guesses is better than a single guess. Ask one person to guess how much a cow weighs, and they'll be off by some amount. Ask a hundred people and take the average of their answers, and you'll be off by less. I was intrigued by a claim in this book review that: You can play "wisdom of crowds" in single-player mode. Say you want to know the weight of a cow. Then take a guess. Now throw your guess out of the window, and take another guess. Finally, compute the average of your two guesses. The claim is that this average is better than your individual guesses. This is spooky. We talk a lot about how to make accurate predictions here - and you can improve your accuracy on anything just by guessing twice and averaging, no additional knowledge required? It's like God has handed us a creepy cow-weight oracle. I wanted to test this myself, so I included some relevant questions in last year's ACX Survey:

https://astralcodexten.substack.com/p/mostly-skeptical-thoughts-on-the People worry about chatbot propaganda. The simplest concern is that you could make chatbots write disinformation at scale. This has created a cottage industry of AI Trust And Safety people making sure their chatbot will never write arguments against COVID vaccines under any circumstances, and a secondary industry of journalists writing stories about how they overcame these safeguards and made the chatbots write arguments against COVID vaccines. But Alex Berenson already writes arguments against COVID vaccines. He's very good at it, much better than I expect chatbots to be for many years. Most people either haven't read them, or have incidentally come across one or two things from his years-long corpus. The limiting factor on your exposure to arguments against COVID vaccines isn't the existence of arguments against COVID vaccines. It's the degree to which the combination of the media's coverage decisions and your viewing habits causes you to see those arguments. A million mechanical Berensons churning out a million times the output wouldn't affect that; even one Berenson already churns out more than most people ever read.

Basically the same as 2022 - this is just a reminder to start working on entries https://astralcodexten.substack.com/p/book-review-contest-rules-2023 Sure, this seemed to go well last the last few times, let's do it again. Write a review of a book. Any book you like - most past winners have been nonfiction, but maybe you can change that! There's no official word count requirement, but previous finalists and winners were often between 2,000 and 10,000 words. There's no official recommended style, but check the style of last year's finalists and winners or my ACX book reviews (1, 2, 3) if you need inspiration. Please limit yourself to one entry per person or team. Then send me your review through this Google Form. The form will ask for your name, email, the title of the book, and a link to a Google Doc. The Google Doc should have your review exactly as you want me to post it if you're a finalist. DON'T INCLUDE YOUR NAME OR ANY HINT ABOUT YOUR IDENTITY IN THE GOOGLE DOC ITSELF, ONLY IN THE FORM. I want to make this contest as blinded as possible, so I'm going to hide that column in the form immediately and try to judge your docs on their merit. (does this mean you can't say something like "This book about war reminded me of my own experiences as a soldier" because that gives a hint about your identity? My rule of thumb is - if I don't know who you are, and the average ACX reader doesn't know who you are, you're fine. I just want to prevent my friends or Internet semi-famous people from getting an advantage. If you're in one of those categories and think your personal experience would give it away, please don't write about your personal experience.) PLEASE MAKE SURE THE GOOGLE DOC IS UNLOCKED AND I CAN READ IT. By default, nobody can read Google Docs except the original author. You'll have to go to Share, then on the bottom of the popup click on "Restricted" and change to "Anyone with the link". If you send me a document I can't read, I will probably disqualify you, sorry. First prize will get at least $2,500, second prize at least $1,000, third prize at least $500; I might increase these numbers later on. All winners and finalists will get free publicity (including links to any other works you want me to link to) and free ACX subscriptions. And all winners will get the right to pitch me new articles if they want (nobody ever takes me up on this). Your due date is April 5th. Good luck! If you have any questions, ask them in the comments. And remember, the form for submitting entries is here.

https://astralcodexten.substack.com/p/response-to-alexandros-contra-me I. In November 2021, I posted Ivermectin: Much More Than You Wanted To Know, where I tried to wade through the controversy on potential-COVID-drug ivermectin. Most studies of ivermectin to that point had found significant positive effects, sometimes very strong effects, but a few very big and well-regarded studies were negative, and the consensus of top academics and doctors was that it didn't work. I wanted to figure out what was going on. After looking at twenty-nine studies on a pro-ivermectin website's list, I concluded that a few were fraudulent, many others seemed badly done, but there were still many strong studies that seemed to find that ivermectin worked. There were also many other strong studies that seemed to find that it didn't. My usual heuristic is that when studies contradict, I trust bigger studies, more professionally done studies, and (as a tiebreaker) negative studies - so I leaned towards the studies finding no effect. Still, it was strange that so many got such impressive results.

https://astralcodexten.substack.com/p/mantic-monday-1302023 One million Metaculi, fake stocks, scandal markets again Happy One Millionth Prediction, Metaculus Metaculus celebrated its one millionth user forecast with a hackathon, a series of talks, and a party: This was a helpful reminder that Metaculus is a real organization, not just a site I go to sometimes to check the probabilities of things. The company is run remotely; catching nine of them in a room together was a happy coincidence. Although I think it still relies heavily on grants, Metaculus' theoretical business model is to create forecasts on important topics for organizations that want them ("partners") - so far including universities, tech companies, and charities. A typical example is this recent forecasting tournament on the spread of COVID in Virginia, run in partnership with the Virginia Department of Health and the University of Virginia Biocomplexity Institute (this year's version here). The main bottleneck is interest from policy-makers, which they're trying to solve both through product improvement and public education. In December, Metaculus' Director of Nuclear Risk, Peter Scoblic, published an article in Foreign Affairs magazine about forecasting's "struggle for legitimacy" in the foreign policy world. It's paywalled, but quoting liberally:

https://astralcodexten.substack.com/p/janus-simulators This post isn't exactly about AI. But the first three parts will be kind of technical AI stuff, so bear with me. I. The Maskless Shoggoth On The Left Janus writes about Simulators. In the early 2000s, the early AI alignment pioneers - Eliezer Yudkowsky, Nick Bostrom, etc - deliberately started the field in the absence of AIs worth aligning. After powerful AIs existed and needed aligning, it might be too late. But they could glean some basic principles through armchair speculation and give their successors a vital head start. Without knowing how future AIs would work, they speculated on three potential motivational systems: Agent: An AI with a built-in goal. It pursues this goal without further human intervention. For example, we create an AI that wants to stop global warming, then let it do its thing. Genie: An AI that follows orders. For example, you could tell it "Write and send an angry letter to the coal industry", and it will do that, then await further instructions. Oracle: An AI that answers questions. For example, you could ask it "How can we best stop global warming?" and it will come up with a plan and tell you, then await further questions. These early pioneers spent the 2010s writing long scholarly works arguing over which of these designs was safest, or how you might align one rather than the other. In Simulators, Janus argues that language models like GPT - the first really interesting AIs worthy of alignment considerations - are, in fact, none of these things. Janus was writing in September 2022, just before ChatGPT. ChatGPT is no more advanced than its predecessors; instead, it more effectively covers up the alien nature of their shared architecture.

https://astralcodexten.substack.com/p/you-dont-want-a-purely-biological CONTENT NOTE: This essay contains sentences that would look bad taken out of context. In the past, I've said "PLEASE DON'T TAKE THIS OUT OF CONTEXT" before or after these, but in the New York Times' 2021 article on me, they just quoted the individual sentence out of context without quoting the "PLEASE DON'T TAKE THIS OUT OF CONTEXT" statement following it. To avoid that, I will be replacing spaces with the letter "N", standing for "NOT TO BE TAKEN OUT OF CONTEXT". If I understand journalistic ethics correctly, they can't edit the sentence to remove the Ns - and if they kept them, people would probably at least wonder what was up.

https://astralcodexten.substack.com/p/who-predicted-2022 Winners and takeaways from last year's prediction contest Last year saw surging inflation, a Russian invasion of Ukraine, and a surprise victory for Democrats in the US Senate. Pundits, politicians, and economists were caught flat-footed by these developments. Did anyone get them right? In a very technical sense, the single person who predicted 2022 most accurately was a 20-something data scientist at Amazon's forecasting division. I know this because last January, along with amateur statisticians Sam Marks and Eric Neyman, I solicited predictions from 508 people. This wasn't a very creative or free-form exercise - contest participants assigned percentage chances to 71 yes-or-no questions, like "Will Russia invade Ukraine?" or "Will the Dow end the year above 35000?" The whole thing was a bit hokey and constrained - Nassim Taleb wouldn't be amused - but it had the great advantage of allowing objective scoring.

https://astralcodexten.substack.com/p/acx-survey-results-2022 Thanks to the 7,341 people who took the 2022 Astral Codex Ten survey. See the questions for the ACX survey See the results from the ACX Survey (click "see previous responses" on that page I'll be publishing more complicated analyses over the course of the next year, hopefully starting later this month. If you want to scoop me, or investigate the data yourself. you can download the answers of the 7000 people who agreed to have their responses shared publicly. Out of concern for anonymity, the public dataset will exclude or bin certain questions. If you want more complete information, email me and explain why, and I'll probably send it to you. Download the public data (.xlsx, .csv) If you're interested in tracking how some of these answers have changed over time, you might also enjoy reading the 2020 survey results. 1 I don't think I can Google Forms only present data from people who agreed to make their responses public, so I've deleted everything identifiable on the individual level, eg your written long response answers. Everything left is just things like "X% of users are Canadian" or "Y% of users have ADHD". There's no way to put these together and identify an ADHD Canadian, so I don't think they're privacy relevant. If you notice anything identifiable on the public results page, please let me know. 2 There will be a few confusing parts. I added some questions halfway through, so they will have fewer responses than others. On the "What Event Led To Your Distrust?" question, I added new multiple choice responses halfway through, so they will incorrectly appear less popular than the other responses. I think that is the only place I did that, but you can email me if you have any questions. 3 I deleted email address (obviously), some written long answers, some political questions that people might get in trouble for answering honestly, and some sex-related questions. I binned age to the nearest 5 years and deleted the finer-grained ethnicity question. I binned all incomes above $500,000 into "high", and removed all countries that had fewer than ten respondents (eg if you said you were from Madagascar, it would have made you identifiable, so I deleted that). If you need this information for some reason, email me. Subscribe to Astral Codex Ten