Astral Codex Ten

TIME

16 min read

TIME

Article

TIME is a recurring publication in the Astral Codex Ten archive, appearing 6 times across 6 issues between May 10, 2021 and November 26, 2025. The archive places it in contexts such as “makes a fascinating contrast with the TIME article on MRAs linked above”; “From TIME”; “TIME: Looking For Love? Let The Market Decide”. It most often appears alongside 9/11, America, Google.

Metadata

  • Category: Publications
  • Mention count: 6
  • Issue count: 6
  • First seen: May 10, 2021
  • Last seen: November 26, 2025

Appears In

Source Context

Recovered passages from the original issue text. When the raw archive preserved outbound links inside the source passage, they are listed directly under the quote.

The Rise And Fall Of Online Culture Wars
May 10, 2021 · Original source
When was the last time you heard people argue about "creeps", "nice guys", or "friendzoning"? Mansplaining? #NotAllMen? MRAs and PUAs? If you're in your early 20s, you might not even know what half these terms mean; if you're older than that, you’ll remember them with a sort of cold dread. But they're gone now - you'd have more luck looking for recent discourse about Osama bin Laden. Nor has some some other gender discourse arisen to replace them. Everyone just stopped caring and moved on to race.
I'm not saying there's literally only one thing the Internet gets in fights about at any given time. The Internet fights about lots of things. But intuitively it feels like there's kind of a power law distribution where one topic clearly outstrips the others - maybe not winner-take-all, but at least winner-take-most. I think you could describe the last twenty years of Internet history as going through three phases - one dominated by religion, one dominated by gender, and now one dominated by race. The race phase seems to have peaked in 2018 and started declining, before being given new life by George Floyd and BLM. The Google Trends results raise the tantalizing possibility that racial issues can’t keep increasing forever. They could eventually crash the same way religious and gender issues did (probably to be replaced by something else even more divisive and awful).
A warning: I was mostly sympathetic to Internet atheism, but mostly unsympathetic to Internet feminism. I think these histories are easier to write from a sympathetic position - any study of Internet culture is basically a study of crazy people, and the failure mode is to point and laugh at them without looking for real understanding. Sometimes pointing and laughing is unavoidable (the New Atheists probably could have done without the Malachi 2:3-related-merchandise) but I think it should be tempered by an attempt at charity. All I can do is try my hardest, and trust readers to keep me honest if I screw up.
Inline links: Malachi 2:3-related-merchandise
Long COVID: Much More Than You Wanted To Know
September 02, 2021 · Original source
First, people with severe COVID that lands them in the ICU have long-lasting symptoms in multiple organ systems. This isn't surprising, and should be considered in the context of post-ICU syndrome. Basically, if anything makes you sick enough to land in the ICU, your body is going to be pretty scarred by the illness (and maybe also by the inevitable side effects of intensive care), and this will last a long time and cause many problems. EG if you’re bedridden for many weeks, your muscles waste away, and then it takes a long time for them to recover and you feel weak and fragile until you do. Or if your lungs stop working and you need mechanical ventilation, your lungs might be pretty weak for a while, and other parts of your body might not get quite the amount of oxygen they’re used to and might get damaged in a way that takes a long time to recover. There’s a similar problem where if you are sufficiently old and frail, any illness will take you down a level of functioning and you might not be able to get up a level again. See for example this article discussing how about 1/5 of elderly flu patients have “persistent functional decline” and may never regain their pre-flu level of functioning.
Inline links: post-ICU syndrome, this article
Second, even in young people with milder cases, COVID can sometimes cause lung damage. If you get lung damage, you’ll have at least breathing problems, and maybe other problems. Your lungs will probably heal eventually, but some kinds of lung healing cause permanent scarring; this can present as shortness of breath on exertion, or become a problem later after other lung injuries.
Inline links: cause lung damage, permanent scarring
Fifth, maybe some long COVID is psychosomatic. People hate when doctors bring up the possibility of psychosomatic conditions, and I won’t deny that we tend to overuse the “psychosomatic” diagnosis like it’s going out of style - but some things really are psychosomatic. Chronic Lyme disease (“Long Lyme” rolls off the tongue nicely) is basically universally considered 100% psychosomatic by the medical establishment, although now that I’m thinking about it I wonder if maybe we should be less sure. Lots of people act like psychosomatic = not a real problem. Unfortunately, having a symptom for psychosomatic reasons sucks just as much as having it for any other reason. Sometimes it sucks more, because nobody takes you seriously. I’ll discuss the argument around psychosomatic symptoms more later.
Inline links: Chronic Lyme disease
24
February 20, 2024 · Original source
How many residents will live in Prospera, a new special economic zone in Honduras, on Jan 1, 2026? Answer: 600 (80% confidence interval 100-2,000) This seems like a good guess (except that my confidence interval would have included zero because there’s a 20%+ chance that it gets shut down). So overall its forecasts seem pretty impressive. But I was concerned by its reasoning even in some of the questions it got “right”. For example, the Nikki Haley question tried to get a base rate by asking what percent of elections Haley had won before, and found she had won 71% of them - these were mostly elections for South Carolina governor. You can see what the AI is trying to do - but it’s not going to work. Then it got confused and read a lot of news stories about how she’s currently losing the 2024 presidential election, and seemed to think they were about 2028. So either the AI only got a reasonable probability by coincidence, or it was testing many different strategies, throwing out the useless ones, and updating only on the useful ones, in a way that was kind of opaque to the casual reader. Still, if the company says it beats most human forecasters, this doesn’t seem totally impossible based on what I’ve seen. And that would be exciting! An AI that can generate probabilistic forecasts for any question seems like in some way a culmination of the rationalist project. And if you can make something like this work, it doesn’t sound too outlandish that you could apply the same AI to conditional forecasts, or to questions about the past and present (eg whether COVID was a lab leak). I would be most excited if at some point this graduated from its geopolitical focus and was able to answer questions on any topic (eg “what is the chance that Astral Codex Ten gains paid subscribers this year?”), maybe if the questioner gives it links or feeds it some of the appropriate information. FutureSearch is run by a team formerly from Metaculus, including former Metaculus CTO (and Google internal prediction market veteran) Dan Schwarz. They’re looking for potential clients and/or investors; if you’re interested, email hello@futuresearch.ai. Vitalik On AI Prediction Markets Vitalik Buterin, Ethereum-founder-turned-cryptocurrency-public-intellectual, has a blog post on The Promise And Challenge Of Crypto + AI Applications. One of them is a prediction market. He writes: Prediction markets have been a holy grail of epistemics technology for a long time; I was excited about using prediction markets as an input for governance ("futarchy") back in 2014, and played around with them extensively in the last election as well as more recently. But so far prediction markets have not taken off too much in practice, and there is a series of commonly given reasons why: the largest participants are often irrational, people with the right knowledge are not willing to take the time and bet unless a lot of money is involved, markets are often thin, etc. One response to this is to point to ongoing UX improvements in Polymarket or other new prediction markets, and hope that they will succeed where previous iterations have failed. After all, the story goes, people are willing to bet tens of billions on sports, so why wouldn't people throw in enough money betting on US elections or LK99 that it starts to make sense for the serious players to start coming in? But this argument must contend with the fact that, well, previous iterations have failed to get to this level of scale (at least compared to their proponents' dreams), and so it seems like you need something new to make prediction markets succeed. And so a different response is to point to one specific feature of prediction market ecosystems that we can expect to see in the 2020s that we did not see in the 2010s: the possibility of ubiquitous participation by AIs. AIs are willing to work for less than $1 per hour, and have the knowledge of an encyclopedia - and if that's not enough, they can even be integrated with real-time web search capability. If you make a market, and put up a liquidity subsidy of $50, humans will not care enough to bid, but thousands of AIs will easily swarm all over the question and make the best guess they can. The incentive to do a good job on any one question may be tiny, but the incentive to make an AI that makes good predictions in general may be in the millions. Note that potentially, you don't even need the humans to adjudicate most questions: you can use a multi-round dispute system similar to Augur or Kleros, where AIs would also be the ones participating in earlier rounds. Humans would only need to respond in those few cases where a series of escalations have taken place and large amounts of money have been committed by both sides. This is a powerful primitive, because once a "prediction market" can be made to work on such a microscopic scale, you can reuse the "prediction market" primitive for many other kinds of questions: Is this social media post acceptable under [terms of use]?
Inline links: FutureSearch, hello@futuresearch.ai, The Promise And Challenge Of Crypto + AI Applications, back in 2014, played around with them extensively, more recently, Polymarket, tens of billions on sports, LK99, Augur
(this would work even better if you could add some humans in the loop who made money by betting against the forecasting bots on the questions where common sense tells you they’re farthest off - but then you’d have to wait human timescales instead of seconds, and pay human wages instead of peanuts)
I’m including it in this post because . . . well, you know how sometimes Christian apologists say that the Gospel writers must have been telling the truth, because they included parts of the story that were embarrassing to them and to their fledgling movement? I’m including this video so future historians will know I must be telling the truth about everything:
Inline links: they included parts of the story that were embarrassing to them
Your Book Review: Nine Lives
September 13, 2024 · Original source
Born in 1978, he was 16 when he left the comfort of his Saudi home, learned to fire a mortar, and fought in the battles of the Bosnian War. He and two friends ran a million dollar fraudulent charity to smuggle supplies to the Chechens when he was 18. He was 19 when he swore an oath of allegiance in front of Osama bin Laden, and started making chemical weapons. He was 20 when he got disillusioned with al-Qaeda, left, got caught by the Qatari secret police and became a British informant. He was 24 when he unraveled a plot to release poison gas in the New York subway. And by the time he was 28, due to an embarrassingly stupid leak from the American intelligence agencies, his spying career was over and he was a man in hiding.
(The jungle war in the Philippines sounded cool in the section title, but his brief stint there at 18 is actually one of the least exciting stories of his life: it was mostly a frozen conflict and the jihadists spent their time playing beach volleyball.)
The book is also a real page-turner, a spy novel in real life. I will share the most interesting things I learned from this book, but for all the adventure stories, read the original, I really enjoyed it more than most novels.
Inline links: read the original
Your Review: The Synaptic Plasticity and Memory Hypothesis
September 12, 2025 · Original source
First a rope was placed around the neck of the captive so that he might not escape; at night the rope was tied to the hammock in which the captive slept. Straps that were not removed were placed above and below the knees. The captives were given women, who guarded them and also slept with them. These women were high-status daughters and sisters of chiefs; they were unmarried and sometimes gave birth to the child of a captive. Some of the captives might be held for a period of time until corn was planted and new large clay vessels—for drink and cooking flesh—were made. Guests were invited to the ceremony, and they often arrived eight to fifteen days in advance of it. A special small house was erected, with no walls but with a roof, in which the captives were placed with women and guards two or three days before the ceremony. In the other houses, feathers were prepared for a headdress or for body ornamentation, and inks were made for tattoos. Women and girls prepared fifty to one hundred vats of fermented manioc beer. Then, when all was ready, they painted the victim’s face blue, mounted a headdress of wax covered with feathers on him, and wound a cotton cord around his waist. The guests began to drink in the afternoon and continued all through the night. At dawn, the one who was to do the killing came out with a long, painted wooden club and smashed the captive on the head, splitting it open. The attacker then withdrew for eight to fifteen days of abstinence while the others ate the cooked flesh of the captive and finished all of the drink made for the occasion.
His final deception was to talk his way onto a French ship, the Catherine, in late 1554. After four months crossing the Atlantic, Europe was in sight, and he was free. He later wrote a bestselling memoir about his time among the Tupinambá, whose English translation was hilariously titled True History: An Account of Cannibal Captivity in Brazil.
I had a hard time finding support for this claim elsewhere. The closest I could find was some discussion by Neil Whitehead, an anthropologist and one of the translators of a recent English version of Staden’s memoir, regarding possible motives for Tupinambá cannibalism rituals beyond revenge. He writes:
Inline links: discussion
Why AI Safety Won't Make America Lose The Race With China
November 26, 2025 · Original source
Compute: America is far ahead. We have better chips (thanks, NVIDIA) and can produce many more of them (thanks, TSMC). Our recent capex boom, where companies like Google and Microsoft spend hundreds of billions of dollars on data centers, has no Chinese equivalent. By the simplest measure - total FLOPs on each sides - we have 10x as much compute as China, and our advantage is growing every day. A 10x compute advantage corresponds to about a 1-2 year time advantage, or an 0.5 - 1 generation advantage (eg GPT-4 to GPT-5).
These are relatively cheap asks. For example, the evaluation to see whether AIs can hack infrastructure will require hiring people who can conduct the evaluation, allocating compute to the evaluation, etc. But on the scale of an AI training run, the sums involved are tiny. Currently, two nonprofits - METR and Apollo Research - do similar tests on publicly-available models. I estimate their respective budgets at $5 million and $15 million per year. Nonprofits can always pay lower salaries than big companies, so it may cost more for OpenAI to replicate their work - for the sake of argument, $25 million. Meanwhile, the likely cost to train GPT-6 will probably be about $25 - $75 billion, with a b. So the safety testing might increase the total cost by 1/1000th. I asked some people who work in AI labs whether this seemed right; they said that most of the cost would be in complexity, personnel, and delay, and suggested an all-things-considered number ten times higher - 1% of training costs.
Inline links: $15 million, $25, $75
All of this lobbying has paid off: the administration keeps proposing changing the rules to allow direct chip sales to China. So far cooler heads have prevailed each time, but the deal keeps popping back onto the table. NVIDIA tries to argue that the models being proposed for export are only second-rate chips that won’t affect the compute balance, but this is false - last month’s talks involved the most price-performant chip on the market. Here’s IFP’s calculation for how caving on this issue would affect the AI race:
Inline links: keeps proposing changing the rules to allow direct chip sales to China, deal keeps popping back onto the table, the most price-performant chip on the market