Category: Uncategorized
“Some Economics of Artificial Super Intelligence”
I promised to pass along serious models of pending AI doom, especially if they are peer-reviewed or at least headed for such. The AI doomer types still are dropping the ball on this, but one economist has made a contribution and so here it is from Henry A. Thompson:
Conventional wisdom holds that a misaligned artificial superintelligence (ASI) will destroy humanity. But the problem of constraining a powerful agent is not new. I apply classic economic logic of interjurisdictional competition, all-encompassing interest, and trading on credit to the threat of misaligned ASI. Using a simple model, I show that an acquisitive ASI refrains from full predation under surprisingly weak conditions. When humans can flee to rivals, inter-ASI competition creates a market that tempers predation. When trapped by a monopolist ASI, its “encompassing interest” in humanity’s output makes it a rational autocrat rather than a ravager. And when the ASI has no long-term stake, our ability to withhold future output incentivizes it to trade on credit rather than steal. In each extension, humanity’s welfare progressively worsens. But each case suggests that catastrophe is not a foregone conclusion. The dismal science, ironically, offers an optimistic take on our superintelligent future.
You may or may not agree, but as usual the point is to build out a literature, not to regard any single paper as the final word. Via the excellent Joy Buchanan.
My first trip to Tokyo
To continue with the biographical segments:
My first trip to Tokyo was in 1992. I was living in New Zealand at the time, and my friend Dan Klein contacted me and said “Hey, I have a work trip to Tokyo, do you want to meet me there?” And so I was off, even though the flight was more of a drag than I had been expecting. It is a long way up the Pacific.
Narita airport I found baffling, and it was basically a two hour, multi-transfer trip to central Tokyo. Fortunately, a Japanese woman was able to help us make the connections. I am glad these days that the main flights come into Haneda.
(One Japan trip, right before pandemic, I decided to spend a whole day in Narita proper. Definitely recommended for its weirdness. Raw chicken was served in the restaurants, and it felt like a ghost town except for some of the derelicts in the streets. This experience showed me another side of Japan.)
We stayed in a business hotel in Ikebukuro, a densely populated but not especially glamorous part of Tokyo. It turned out that was a good way to master the subway system and also to get a good sense of how Tokyo was organized. I had to one-shot memorize the rather complicated footpath from the main subway station to the hotel, which had been chosen by my friend’s sponsors. As we first emerged from the subway station, we had, getting there the first time, to ask two Japanese high schoolers to help us find the way. They spoke only a few words of English, but we showed them the address in Japanese and they even carried our bags for us, grunting “Hai!” along the way, giving us a very Japanese experience.
In those days very little English was spoken in Tokyo, especially outside a few major areas such as Ginza. You were basically on your own.
I recall visiting the Sony Center, which at the time was considered the place to go to see new developments in “tech.” I marveled at the 3-D TV, and realized we had nothing like it. I felt like I was glimpsing the future, but little did I know the technology was not going anywhere. Nor for that matter was the company. Here is Noah, wanting the Japanese future back.
Most of all, Tokyo was an extreme marvel to me. I felt it was the single best and most interesting place I had visited. Everywhere I looked — even Ikebukuro — there was something interesting to take note of. The plastic displays of food in the windows (now on the way out, sadly) fascinated me. The diversity, order, and package wrapping sensibilities of the department stores were amazing. The underground cities in the subways had to be seen to be believed (just try emerging from Shinjuku station and finding the right exit). The level of dress and stylishness and sophistication was extreme, noting I would not say the same about Tokyo today. This was not long after the bubble had burst, but the city still had the feel of prosperity. Everything seemed young and dynamic.
I also found Tokyo affordable. The reports of the $2,000 melon were true, but the actual things you would buy were somewhat cheaper than in say New York City. It was easy to get an excellent meal for ten dollars, and without much effort. My hotel room was $50 a night. The subway was cheap, and basically you could walk around and look at things for free. The National Museum was amazing, one of the best in the world and its art treasures cannot, in other forms, readily be seen elsewhere.
Much as I like Japanese food, I learned during this trip that I cannot eat it many meals in a row. This was the journey where I realized Indian food (!) is my true comfort food. Tokyo of course has (and had) excellent Indian food, just as it has excellent food of virtually every sort. I learned a new kind of Chinese food as well.
The summer heat did not bother me. I also learned that Tokyo is one of the few cities that is better and more attractive at night.
I recall wanting to buy a plastic Godzilla toy. I walked around the proper part of town, and kept on asking for Godzilla. I could not figure out why everyone was staring at me like I was an idiot, learning only later that the Japanese say “Gojira.” So in a pique of frustration, I did my best fire-breathing, stomping around, “sound like a gorilla cry run backwards through the tape” imitation of Godzilla. Immediately a Japanese man excitedly grabbed me by the hand, walked me through some complicated market streets, and showed me where I could buy a Godzilla, shouting “Gojira, Gojira, Gojira!” the whole time.
I came away happy.
My side trip, by the way, was to the shrines and temples of Kamakura, no more than an hour away but representing another world entirely. Recommended to any of you who are in Tokyo with a day to spare.
Now since that time, I’ve never had another Tokyo trip quite like that one. These days, and for quite a while, the city feels pretty normal to me, rather than like visiting the moon. Fluent English is hard to come by, but most people can speak some English and respond to queries. You can translate and get around using GPS, AI, and so on. The city is much more globalized, and other places have borrowed from its virtues as well.
Looking back, I am very glad I visited Tokyo in 1992. The lesson is that you can in fact do time travel. You do it by going to some key places right now.
Thursday assorted links
2. John Cochrane on causal identification.
3. What are the density trends in major cities around the world?
4. “You Can Now Invest In A Hedge Fund Dedicated To Hermès Bags.”
5. What does it mean for an experiment to be beautiful?
8. New Yorker fact checkers confirm that AI song really did hit #1 on the C&W charts.
9. Here is the 5.1 Thinking model on different kinds of humor in Woody Allen movies, and then a follow-up question on the influence of Bergman. Impressive, as is the model more generally.
My Conversation with the excellent Donald S. Lopez Jr.
Here is the audio, video, and transcript. Here is part of the episode summary:
Tyler and Donald discuss the Buddha’s 32 bodily marks, whether he died of dysentery, what sets the limits of the Buddha’s omniscience, the theological puzzle of sacred power in an atheistic religion, Buddhism’s elaborate system of hells and hungry ghosts, how 19th-century European atheists invented the “peaceful” Buddhism we know today, whether the axial age theory holds up, what happened to the Buddha’s son Rahula, Buddhism’s global decline, the evidently effective succession process for Dalai Lamas, how a guy from New Jersey created the Tibetan Book of the Dead, what makes Zen Buddhism theologically unique, why Thailand is the wealthiest Buddhist country, where to go on a three-week Buddhist pilgrimage, how Donald became a scholar of Buddhism after abandoning his plans to study Shakespeare, his dream of translating Buddhist stories into new dramatic forms, and more.
Excerpt:
COWEN: Fire is a central theme in Buddhism, right?
LOPEZ: Well, there are hot hells, and there are also cold hells. Fire comes up, really, in the idea of nirvana. Where we see the fire, I think most importantly, philosophically, is the idea of where did the Buddha go when he died? He was not reborn again. They say it’s really just like a flame going out, that is, the flame ends. Where did the fire go? Nowhere, that is, the wood that was producing the flame is all burned up, and you just end. Nirvana is not a place. It’s a state of extinction or what the Buddhists call cessation.
COWEN: What role does blood sacrifice play in Buddhism?
LOPEZ: Well, it’s not supposed to perform any role. There’s no blood sacrifice in Buddhism.
COWEN: No blood sacrifice. How about wrathful deities?
LOPEZ: Wrathful deities — there’re a lot, yes.
COWEN: Then we’re back to supernatural. Again, this gets to my central confusion. It’s atheistic, but there’s some other set of principles in the universe that generate wrathful deities, right?
LOPEZ: Wrathful deities are beings who were humans in one lifetime, animals in another, and born as wrathful deities in another lifetime. Everyone is in the cycle of rebirth. We’ve all been wrathful deities in the past. We’ll be wrathful deities in the future unless we get out soon. It’s this universe of strange beings, all taking turns, shape-shifting from one lifetime to the next, and it goes on forever until we find the way out.
COWEN: Are they like ghosts at all — the wrathful deities?
LOPEZ: There’s a whole separate category of ghosts. The ghosts are often called — if we look at the Chinese translation — hungry ghosts. The ghosts are beings who suffer from hunger and thirst. They are depicted as having distended bellies. They have these horrible sufferings that when they drink water, it turns into molten lead. They’ll eat solid food — it turns into an arrow or a spear. Constantly seeking food, constantly being frustrated, and they appear a lot in Buddhist text. One of the jobs of Buddhist monks and nuns is to feed the hungry ghosts.
COWEN: Is it a fundamental misconception to think of Buddhism as a peaceful religion?
And:
COWEN: …If one goes to Borobudur in Java — spectacular, one of the most amazing places to see in the world.
LOPEZ: Absolutely.
COWEN: We read that it was abandoned. It wasn’t even converted into a tourist site or a place where you would sell things. Why would you just toss away so much capital structure?
LOPEZ: I think it just got overgrown by the jungle. I think that people were not going there. There were no Buddhist pilgrims coming. The populace converted to Islam mostly, and it just fell into decline, just to be revived in the 19th, 20th century.
COWEN: Turn it into a candy store or something! It just seems capital maintenance occurs across other margins. The best-looking building you have — one of the best-looking in the world — is forgotten. Don’t you find that paradoxical?
Definitely recommended, interesting throughout and I learned a great deal doing the prep. One of my favorite episodes of this year. And I am happy to recommend all of Donald’s books on Buddhism.
Prediction markets in everything? Tariff refund edition
Oppenheimer changed its terms from offers earlier this year. The firm said it would consider bids starting at 20 percent per refund claim pertaining to “reciprocal” or IEEPA tariffs and 10 percent for tariffs tied to fentanyl.
Gabriel Rodriguez, the president and co-founder of A Customs Brokerage, in Doral, Fla., and a recipient of several emails from Oppenheimer, said he believed Oppenheimer was offering to pay the equivalent of 80 cents on the dollar per claim.
Here is more from the NYT, via Amy.
Wednesday assorted links
1. Marginal product. And more.
2. On Richard II.
3. Brief comments on stablecoins.
4. Decorating with information density (NYT).
5. Susan Athey and Fiona Scott Morton on the economics of AI. They actually have something new to say.
Tuesday assorted links
2. Road network of the Roman empire.
3. NIMBYs will pay to keep low income people away. Job market paper from Helena Pedrotti of NYU.
4. Kennedy Center update (NYT). And the Washington National Opera might leave the Kennedy Center.
5. Decoupling dollar and Treasury privilege.
6. In a mock trial, the AIs acquit a teen who was ruled guilty by the judge.
Old school workplace feminization
We investigate whether consequential decisions made by judges are impacted by the gender composition of these judges’ peer group. Using the universe of decisions on juvenile defendants in each courthouse in a Southern state over 15 years, we estimate two-way fixed effects models leveraging random assignment of cases to judges and variations in judge peer composition generated by judicial turnover. The results show that an increase in the proportion of female peers in the courthouse causes a rise in individual judges’ propensity to incarcerate, and an increase in prison time. This effect is driven by the behavior of female judges. We examine the sensitivity of our findings to heterogeneous-robust difference-in-differences estimators for continuous and nonabsorbing treatments.
Here is the full article by Ozkan Eren and Naci Mocan, tekl.
*Marked by Time*
The author is Robert J, Sampson, and the subtitle is How Social Change has Transformed Crime and the Life Trajectories of Young Americans, from Harvard University Press. Excerpt:
…[for part of Chicago]..the chance of being arrested in life among people born in the mid-1980s is more than double that of those born just a decade later, in the mid-1990s. This large arrest inequality does not arise from early-life individual, family, or local neighborhood characteristics. It arises from the larger and highly divergent socio-historical contexts in which the children grew through adolescence into adulthood.
The particular story focuses on guns, death, and lead exposure, though I wonder whether the in-sample implied elasticities are validated out of sample. Nonetheless an interesting book.
Monday assorted links
2. “Risk of myocarditis is higher after Covid infection than after vaccine.” For young people, yup.
3. Claims about Tokyo’s train system. Princeton job market paper.
4. Matt Y. on the game theory of some of the Democrats conceding. And thoughts from Sam Stein.
5. Japanese buzzwords of the year?
6. Only five countries have no Swiss citizens living there. It seems it would be more efficient for that number to be higher?
7. The #1 country song is AI-generated?
8. Efforts toward genetically-engineered babies (WSJ). And the “fertility awareness” movement (NYT).
In defense of Schumpeter
Factories of Ideas? Big Business and the Golden Age of American Innovation (Job Market Paper) [PDF]
This paper studies the Great Merger Wave (GMW) of 1895-1904—the largest consolidation event in U.S. history—to identify how Big Business affected American innovation. Between 1880 and 1940, the U.S. experienced a golden age of breakthrough discoveries in chemistry, electronics, and telecommunications that established its technological leadership. Using newly constructed data linking firms, patents, and inventors, I show that consolidation substantially increased innovation. Among firms already innovating before the GMW, consolidation led to an increase of 6 patents and 0.6 breakthroughs per year—roughly four-fold and six-fold increases, respectively. Firms with no prior patents were more likely to begin innovating. The establishment of corporate R\&D laboratories served as a key mechanism driving these gains. Building a matched inventor–firm panel, I show that lab-owning firms enjoyed a productivity premium not due to inventor sorting, robust within size and technology classes. To assess whether firm-level effects translated into broader technological progress, I examine total patenting within technological domains. Overall, the GMW increased breakthroughs by 13% between 1905 and 1940, with the largest gains in science-based fields (30% increase).
That is the job market paper of Pier Paolo Creanza, who is on the market this year from Princeton.
*Violent Saviors*
That is the new William Easterly book, and the subtitle is The West’s Conquest of the Rest. I liked this book very much, but found the title and also book jacket and descriptions misleading. I think of this work as a full-throated examination and study of the classical liberal anti-imperialist tradition. We have been needing such a thing for a long time. It is not that I expected Easterly to be poorly informed, but it amazes me how well he knows this material from a historical point of view. A lengthy (and good) discussion of E.D. Morel!
So recommended, and here’s hoping these traditions find some new legs and less crazy adherents.
Sunday assorted links
1. Farmers’ Almanac will cease publication.
2. Econ history job candidates this year, a list.
3. How an econ job market candidate should maximize his or her LinkedIn profile.
4. “Mamdani is under fire for charging people $13 per beer at his ‘victory party’.”
5. Burn-Murdoch covers Noy and Rao on polarizing media (FT).
Should USG support a 50-year mortgage?
More “affordability” from Trump!? From GPT-5:
Broad, government‑backed 50‑year mortgages would likely lower monthly payments but raise house prices, slow equity build‑up (and raise default risk in downturns), and increase interest‑rate risk in the financial system. As a general affordability policy for the U.S., that’s a poor trade‑off. If used at all, a 50‑year term should be tightly targeted (e.g., to loan modifications or to newly built homes only), not adopted wholesale for new purchases…
In the short run, sellers and incumbent owners capture much of the benefit via higher sale prices; first‑time buyers face higher entry prices…
Here is the whole answer. Albania has the right idea!
The first human arrivals in the New World may have sailed from northeast Asia
The first people to migrate to North America may have sailed from north-east Asia around 20,000 years ago. Experts have argued that prehistoric people in Hokkaido, Japan, used similar stone tools to those later found in North America, and suggest that seafarers may have travelled to the continent during the last ice age, bringing this stone technology with them. This adds weight to the theory that the first Americans arrived much earlier than previously thought.
And:
“By [around 30,000 years ago], Upper Palaeolithic seafarers were using sea-going vessels to access some of the outer islands in the Japanese archipelago … and were capable of negotiating the Kuroshio Current, one of the fastest in the world,” Davis and his colleagues write in their paper, published in October in the journal Science Advances. “This suggests that such experienced seafarers may also have been capable of handling adverse Pacific coastal currents.”
Nonetheless, the team also suggest that the journey could have happened at a much slower pace. In this reconstruction, the prehistoric seafarers gradually followed a route along the Pacific coast.
Here is the full story. I have long wondered why, every now and then, in Japan I see a person who looks a great deal like a “Native American.”