A short note on the importance of fun [Homo ludens]

Tyler Cowen has just interviewed economic historian Sheilagh Ogilvie on a variety of topics:

...the economic impacts of historical pandemics, the “happy story” of the Black Death and why it doesn’t stand up to scrutiny, the history of variolation and how entrepreneurs created vaccination franchises in 18th-century England, why local communities typically managed epidemics better than central authorities, the dastardly nature of medieval guilds, the European marriage pattern and its disputed contribution to economic growth, when sustained economic growth truly began in England, why the Dutch Republic stagnated despite its early success, whether she agrees with Greg Clark’s social mobility hypothesis, her experience and conducting “anthropological fieldwork” on English social customs, the communitarian norms she encountered while living in Germany...

It's all interesting. But, given my current interests, this passage caught my eye:

COWEN: Let’s say an 18-year-old, highly intelligent young woman comes to you. She’s moving to England; she might want to be a professor. What advice do you give to her? From America, let’s say.

OGILVIE: [...] I would say, and this is actually not just England-specific, but changing cultures. When I was 20, I went to live in Germany. I lived there, actually, for quite a long time while I was doing research for my doctorate.

When you initially go to a foreign place, whether it’s Germany — or later, I lived in the Czech Republic for a year — you won’t figure out right away what people in this culture do to feel comfortable and have fun. But you have to have the faith that they do have things that they do, and you need to learn what those things are that they do in this culture to be comfortable with one another and to have a great time.

The sooner you learn that, the happier you’ll be fitting into a new culture. I think that would be my general life advice to anyone who is moving to a new country. Find out what people do for fun, and then start doing it.

I suspect that there is more to that advice than the fact that fun is, well, fun. There's something deeper going on. It is not just that fun is a better way to pass the time than boredom or even work, but that it puts you more fundamentally touch with the people.

Summertime, Sarah "the voice" Vaughan

Lawrence Summers on why universities must resist Trump

Lawrence H. Summers, If Powerful Places Like Harvard Don’t Stand Up to Trump, Who Can? NYTimes, April 3, 2025.

The U.S. government is trying to bludgeon America’s elite universities into submission. [...] The Trump administration’s threats to withdraw billions of dollars in funding are little more than extortion. They must be resisted using all available legal means.

Summers asserts that elite universities do have problems with antisemitism, identity and diversity issues, and "have repeatedly failed to impose discipline and maintain order." Reform is needed. However:

...the Trump administration is not acting in good faith in its purported antisemitism concerns, nor is it following the law in its approach to universities. [...] Title VI of the 1964 Civil Rights Act appropriately allows that federal funding of universities can be made contingent on their avoiding discrimination. But as a recent statement by a group of leading law professors points out, it also protects against this power’s being used to punish critics or curtail academic freedom. Among the law’s requirements are notice periods, hearings, remedies that are narrowly tailored to specific infractions and a 30-day congressional notification before any funding is curtailed.

None of this appears to be part of the Trump administration’s approach to universities.

The White House has not confined its efforts to claims about discrimination. The administration seeks to dictate what universities do on matters ranging from student discipline to academic organization to campus policing.

His penultimate paragraph:

Institutions such as Harvard, the administration’s most recent target, have vast financial resources, great prestige and broad networks of influential alumni. If they do not or cannot resist the arbitrary application of government power, who else can? Without acts of resistance, what protects the rule of law?

Flash!

Perspective on musical neurodynamics

Harding, E.E., Kim, J.C., Demos, A.P. et al. Musical neurodynamics. Nature Reviews Neuroscience. (2025). https://doi.org/10.1038/s41583-025-00915-4

Abstract: A great deal of research in the neuroscience of music suggests that neural oscillations synchronize with musical stimuli. Although neural synchronization is a well-studied mechanism underpinning expectation, it has even more far-reaching implications for music. In this Perspective, we survey the literature on the neuroscience of music, including pitch, harmony, melody, tonality, rhythm, metre, groove and affect. We describe how fundamental dynamical principles based on known neural mechanisms can explain basic aspects of music perception and performance, as summarized in neural resonance theory. Building on principles such as resonance, stability, attunement and strong anticipation, we propose that people anticipate musical events not through predictive neural models, but because brain–body dynamics physically embody musical structure. The interaction of certain kinds of sounds with ongoing pattern-forming dynamics results in patterns of perception, action and coordination that we collectively experience as music. Statistically universal structures may have arisen in music because they correspond to stable states of complex, pattern-forming dynamical systems. This analysis of empirical findings from the perspective of neurodynamic principles sheds new light on the neuroscience of music and what makes music powerful.

Generative AI Chatbot for Mental Health Treatment

Michael V. Heinz, Daniel M. Macklin, Brianna M. Trudeau, et al., Randomized Trial of a Generative AI Chatbot for Mental Health Treatment, The New England Journal of Medicine, VOL. 2 NO. 4, March 27, 2025, DOI: 10.1056/AIoa2400802

Abstract

Background

Generative artificial intelligence (Gen-AI) chatbots hold promise for building highly personalized, effective mental health treatments at scale, while also addressing user engagement and retention issues common among digital therapeutics. We present a randomized controlled trial (RCT) testing an expert–fine-tuned Gen-AI–powered chatbot, Therabot, for mental health treatment.

Methods

We conducted a national, randomized controlled trial of adults (N=210) with clinically significant symptoms of major depressive disorder (MDD), generalized anxiety disorder (GAD), or at clinically high risk for feeding and eating disorders (CHR-FED). Participants were randomly assigned to a 4-week Therabot intervention (N=106) or waitlist control (WLC; N=104). WLC participants received no app access during the study period but gained access after its conclusion (8 weeks). Participants were stratified into one of three groups based on mental health screening results: those with clinically significant symptoms of MDD, GAD, or CHR-FED. Primary outcomes were symptom changes from baseline to postintervention (4 weeks) and to follow-up (8 weeks). Secondary outcomes included user engagement, acceptability, and therapeutic alliance (i.e., the collaborative patient and therapist relationship). Cumulative-link mixed models examined differential changes. Cohen’s d effect sizes were unbounded and calculated based on the log-odds ratio, representing differential change between groups.

Results

Therabot users showed significantly greater reductions in symptoms of MDD (mean changes: −6.13 [standard deviation {SD}=6.12] vs. −2.63 [6.03] at 4 weeks; −7.93 [5.97] vs. −4.22 [5.94] at 8 weeks; d=0.845–0.903), GAD (mean changes: −2.32 [3.55] vs. −0.13 [4.00] at 4 weeks; −3.18 [3.59] vs. −1.11 [4.00] at 8 weeks; d=0.794–0.840), and CHR-FED (mean changes: −9.83 [14.37] vs. −1.66 [14.29] at 4 weeks; −10.23 [14.70] vs. −3.70 [14.65] at 8 weeks; d=0.627–0.819) relative to controls at postintervention and follow-up. Therabot was well utilized (average use >6 hours), and participants rated the therapeutic alliance as comparable to that of human therapists.

Conclusions

This is the first RCT demonstrating the effectiveness of a fully Gen-AI therapy chatbot for treating clinical-level mental health symptoms. The results were promising for MDD, GAD, and CHR-FED symptoms. Therabot was well utilized and received high user ratings. Fine-tuned Gen-AI chatbots offer a feasible approach to delivering personalized mental health interventions at scale, although further research with larger clinical samples is needed to confirm their effectiveness and generalizability. (Funded by Dartmouth College; ClinicalTrials.gov number, NCT06013137.)

This song has been been haunting me for two or three days: Chan Chan

This is the original version:

From the webpage:

The original Buena Vista Social Club album was recorded for World Circuit Records over six days at the vintage EGREM studios in Havana. Released in 1997, the album went on to become a worldwide phenomenon, selling over 8 million copies and contributing to the rise in popularity of Cuban, as well as Latin American, music.

This is a version by Playing for Change:

From the webpage:

This song reached the masses from the Buena Vista Social Club project and I have heard so many variations and versions performed around the world ever since. We decided to create our own version while on a trip recording and filming music in Cuba in 2015. The Legendary Pancho Amat on the Cuban tres along with the incredible piano playing of Roberto Carcasses set the framework for this song, and then once we heard Teté Garcia Caturla sing lead vocals we realized exactly why we do what we do. Listen to how well Cuba, the USA, and the Middle East all get along when the music plays. - Mark Johnson, PFC producer and co-founder

AI & humans, then and now

In “The Evolution of Cognition” (1990) David Hays and I argued that the long-term evolution of human culture flows from the architectural foundations of thought and communication: first speech, then writing, followed by systematized calculation, and most recently, computation. In discussing the importance of the computer, we remark:

One of the problems we have with the computer is deciding what kind of thing it is, and therefore what sorts of tasks are suitable to it. The computer is ontologically ambiguous. Can it think, or only calculate? Is it a brain or only a machine?

The steam locomotive, the so-called iron horse, posed a similar problem for people at Rank 3. It is obviously a mechanism and it is inherently inanimate. Yet it is capable of autonomous motion, something heretofore only within the capacity of animals and humans. So, is it animate or not? Perhaps the key to acceptance of the iron horse was the adoption of a system of thought that permits separation of autonomous motion from autonomous decision. The iron horse is fearsome only if it may, at any time, choose to leave the tracks and come after you like a charging rhinoceros. Once the system of thought had shaken down in such a way that autonomous motion did not imply the capacity for decision, people made peace with the locomotive.

The computer is similarly ambiguous. It is clearly an inanimate machine. Yet we interact with it through language; a medium heretofore restricted to communication with other people. To be sure, computer languages are very restricted, but they are languages. They have words, punctuation marks, and syntactic rules. To learn to program computers we must extend our mechanisms for natural language.

Back then the question was mostly an academic one. That is to say, it had little purchase on the daily lives of most people. Consequently, however intently a relatively small cadre of academics debated the question, it was of relatively little interest to ordinary people.

That changed quite dramatically when, late in November 2022, OpenAI released ChatGPT on the web where anyone with an internet account and a web browser to access it and play with it. Overnight millions did so. The question of whether or not this thing was dead or alive, that is inanimate or animate, mindless or conscious, impressed itself on millions of users. It was no longer an academic question. It was a live question, and to some it was even existential: How long before this, this, this THING, goes rogue and destroys us?

Fortunately, that has not happened. We are all alive to debate the issue. And we do so, using terms that existed long prior to the release of ChatGPT. That’s a problem.

Back in the days when the questions of computer intelligence, of computational minds, and of artificial consciousness were academic, we had no examples of devices whose behavior was phenomenologically problematic. Computers played an inferior game of chess, though that ended in 1997 when IBM’s Deep Blue defeated Gary Kasparov, and were at best halting, clumsy, and relentlessly stupid with language. You could take whatever position you wished about the possibility of artificial intelligence (AI), artificial general intelligence (AGI), a term coined early in the millennium, or even superintelligence, a term popularized by Nick Bostrom’s 2013 book of that title, when it came to actual devices, it was clear that they were not intelligent or conscious.

ChatGPT could “talk,” just like a human being, or so much so that one had to work hard to find a meaningful difference. Many users proceeded as though there were no meaningful difference. Now the question of AI, AGI, or even superintelligence has taken on a different valence. Any chatbot “knows” a wider range of subjects than even the most brilliant and learned of humans. In that specific and limited sense, these things are superintelligent. While no one, so far as I know, has claimed that these chatbots are superintelligent in the fullest sense (as in Bostrom’s Superintelligence) you see the problem. Don’t you?

Just as we don’t know how the human mind, the human brain, works. So we don’t know how these chatbots, these large language models (LLMs) work. Do they work like we do or not? At some level obviously not. Computer hardware is quite different from biological “wetware” (brains). But when we consider function, that we don’t know. As long as we stick to symbolic behavior, the ability to write natural language, and increasingly to speak it, to write computer code, and to worth with mathematics, our ability to distinguish the real, that is, humans, from the artificial, that is, computers, is problematic. Thus the terms, the concepts, we have inherited from the pre-GPT-3 era are no longer adequate to problems we now face.

That is what makes the question of computer intelligence both so urgent and so deeply problematic. For the moment, I’m fond of a formulation Steven Harnad expressed somewhere on the web: The behavior chatbots exhibit is astonishing when you consider the fact that they don’t understand another. Those words are mind, but the though is Harnad’s. But this formulation is no more than a stop gap.

We need new concepts, and new conceptual framework. That’s easier called for than accomplished. The accomplishment will take an intellectual generation.

* * * * *

I’ve made this general point, but at greater length and in different terms in a paper I finished in January of 2023: ChatGPT intimates a tantalizing future; its core LLM is organized on multiple levels; and it has broken the idea of thinking.

Caught in delium [blossoms]

Tyler Cowen interviews Ezra Klein on abundance, kludgeocracy and so much more

Here's the link, and the intro:

What happens when a liberal thinker shifts his attention from polarization to economic abundance? Ezra Klein’s new book with Derek Thompson, Abundance, argues for an agenda of increased housing, infrastructure, clean energy, and innovation. But does abundance clash with polarization—or offer a way through it?

In this conversation, Ezra and Tyler discuss how the abundance agenda interacts with political polarization, whether it’s is an elite-driven movement, where Ezra favors NIMBYism, the geographic distribution of US cities, an abundance-driven approach to health care, what to do about fertility decline, how the U.S. federal government might prepare for AGI, whether mass layoffs in government are justified, Ezra’s recommended travel destinations, and more.

The interview covers a lot of territory. Check it out for yourself. I present two excerpts below:

On the abundance agenda as an elite movement

COWEN: Is the abundance agenda primarily a view of elites? And insofar as it succeeds, it will succeed to the extent that politics is not directly ruled by democratic forces?

KLEIN: I don’t know that I think that’s true, but it’s always a little bit true. In my own head, I think there are two kinds of policy and procedure overhang. One is the kind people actually want. A lot of NIMBYism is popular at the level at which it is happening.

But then, there’s a kind that comes from drift. It doesn’t end up being in the book, but I wrote a Times piece about this. I don’t know if you tracked the story a while back that there was this public toilet being built in Noe Valley in a park — I used to live near this park — that was hooked up for water. The cost estimate was $1.7 million. They released a ribbon cutting for this. [laughs] They came out to announce that they had gotten the money from the State of California to build this $1.7 million toilet.

Then my colleague, Heather Knight, back then at the San Francisco Chronicle, reported on this, and people freaked out. They were not happy that the city had gotten $1.7 million for a toilet. They thought, “Why?” I went and tracked down, how does a toilet get to $1.7 million? One of the justifications from Rec and Parks in SF was, “Look, we’ve built all these other toilets that were $1.6, $1.7 million, and nobody complained about that.”

It’s this baroque process where you have seven, eight, nine agencies. You have all these public comment periods. There are all these rules on the grant proposal, and how that goes out, and what the procurement is, and how you do the bids on the contracts. Nobody asked for that. That’s drift. That is process building on top of itself. That is nobody really having the power to say no or wanting to go through the difficulty of saying no.

I cut that differently from some of the — there are places where people do not want to see an affordable housing complex built down the block. Then there are places where people would actually like to have an affordable public restroom next to the playground where there are kids who — I’ve been through this period of mine very recently — who are not fully potty-trained, playing. They’re not out there hoping that we can add $1.3 million in cost by process; they just don’t really know that it has happened.

COWEN: If the abundance agenda is, to some extent, an elite movement, it seems that high density, for the most part, is bad for elites. In California, very wealthy elites — they want to move to Woodside. In Northern Virginia, they want to move to McLean. Does this mean the NIMBY part of the movement is just never going to get very far? That you can take a bunch of places, say, near metro stops, and allow for somewhat denser housing, but it will stop there.

KLEIN: Is that true, though? I live in New York City now, and my sense is, very wealthy elites — while they have a weekend house or vacation house, possibly in the Hamptons, they live in Manhattan in glass towers, or they’ve started to buy really expensive property in Brooklyn or in —

COWEN: But that’s one state, right? The country as a whole — the elites want the ranch two hours from Houston.

KLEIN: It seems they want both, to me. I feel like, the elites of DC live in McLean — that was true 20 years ago, and it feels less true now. You still live there, but is that really what people do when they get rich in DC now? I feel like they buy expensive DC property because DC got safer, and the food got better, and all the things that everybody knows.

Kludgeocracy and DOGE

COWEN: You know full well Steven Teles’s work on kludgeocracy. We have at least 50 years of the kludge that is just accumulating. Everyone in government knows about this. It’s not some secret. Various administrations have tried to address it. Al Gore did a bit — that was fine, but it didn’t really stop it. I don’t see that there’s any other recipe besides quite a bit of disruption. Again, there’ll be future administrations to sort it out.

Like the New Deal agencies — they weren’t so great to begin with. They didn’t have experience or data or staffing, but over time, pieces fell into place. Maybe the options are just more and more Steven Teles’s kludgeocracy or we take some chances today and do some things that actually hurt. Then over an 8- to 12-year period sort it out with AI and most of all, with future administrations, I don’t really see what the alternative is.

KLEIN: Well, I think there are two questions here. Let’s say there’s a good version of this and a bad version of this. I’ll outline the way I see both. Let me start with a good version of this. If I was saying what I hope the story of this period will ultimately be, I have to describe it as thesis, antithesis, and synthesis, which I think is a little bit of what you’re saying.

My critique of Democrats is, they became culturally process- and bureaucracy-obsessed. They saw the state and inside their own agencies, inside things they, in theory, run, they were anything a lawyer said you had to take as holy writ. No matter how off the wall or stringent the interpretation of what was clearly a looser original statute was, it was careful. Every process had to be followed to a T. Democrats, in theory, the party of government, cannot run government effectively because of some of the reasons you just described and some of the reasons I just described.

Then in comes Donald Trump and Elon Musk, and if they have proven anything, it is that a lot of the things that previous administrations — by the way, Democratic and Republican — treated as inviolable were just not. They weren’t real. They were norms, not rules, norms not laws. It is clear that some of the firings were illegal. Clearly not all of them. It is clear that some of the withholding of money is illegal. My gut is, having seen the five-four decision the other day at the Supreme Court which went against the Trump administration — but it was a five-four decision — it’s not going to be all of it.

There is a lesson that is being taught here by Musk, which I think liberals have to look at very uncomfortably, which is that things that they treated as facts of the system that could not be in any way altered and then used as excuses for low-performing government services for genuinely . . . Probably what Elon Musk is calling waste and what I would call waste are not the same thing, but there is what I would think of as a lot of waste. They allowed a civil service system to emerge and evolve that everybody knows is crazy. Everybody knows it’s crazy.

Here comes Musk and Trump — antithesis. I don’t think they’re trying to make things work. I don’t think it’s zero-based budgeting. I don’t think they’re holding things to a standard. I think they’ve cut off huge amounts of lifesaving work. I think they are creating a lot of risk in parts of the system that could really blow up. I think what they want is control, not a working government.

Authoritarianism in the US and Abroad | Robert Wright & Almut Rochowanski

0:00 Almut’s first-hand acquaintance with authoritarianism
2:00 Trump 2.0’s authoritarian hallmarks
19:13 What should worry us most about Trump 2.0?
25:51 Trump 2.0 vs post-Soviet oligarchy: compare and contrast
32:44 Is Trump an ethno-nationalist?
38:38 Heading to Overtime

The whole discussion is interesting, but I'd like to call your attention to the discussion at 19:13, where Wright poses the question: How worried should we be? That is, how close are we to losing the republic and tipping over into a full-blown authoritarian state, if not in name, but in practice? Her reply (c. 19:58):

Rochowanski: All of this happens on grade, right? Like on a curve like it's already this is already it has already happened in a way. And positionality comes in very much here. Like if you've been a migrant belong to certain groups of migrants in this country this has already been your reality for a very long time, for decades possibly. It's not for nothing that ICE has the reputation that it has. It's just you know until recently this happened to uh migrants of different classes and now it happens to migrants who are at Ivy League graduate programs. That is new, right.

But this is you know in many ways that has already been true. I mean and certainly you know like been there. I mean I remember you know I was already here and already doing this work you know, when Citizens United happened. And like sort of the the howling and gnashing of teeth that that was heard around the time. Saying like this is going to be the end of the republic. And guess what? You know this is going to be the old lead us into the oligarchy and it very much looks like it has? So in a way like you should have been worried a long time ago.

Later, they're discussing Roe vs. Wade (c. 24:37):

Rochowanski: You know like. you really shouldn't do victimhood Olympics and all of that. But in the end you know women still are, sort of our rights are the most disposable of all, the the ones about which one can you know reasonably have discussions about, like they are up for debate and other things are not so much up for debate. Like freedom of speech. But I mean yeah it does certainly like there's something new about it. I don't want to deny that; I don't want to like you know sort of like say, oh nothing to see here. Not at all. But I just, you asked me like should we be worried and my point is you should have been worried a long time.

A selection of recent photos

Superforcasting the future of AI [is the methodology appropriate? – a discussion with Claude 3.7]

Superforcasting is a technique for making predictions about world events that originated in The Good Judgment Project. I’ve known about it for a while and read about it here and there. Yesterday (March 31, 2025) Malcolm Murray published A Superforecaster’s View on AGI in 3 Quarks Daily. I won’t try to summarize the article, but I do want to extract a couple of things from it.

First he distinguishes between input and output definitions of AGI. Judging from his overall usage, he means input to and output from the economy, but he doesn’t explicitly say that. Input definitions attempt to characterize AGI by the cognitive tasks the system can perform. Most debates about AGI are about this. In contrast, output definitions characterize AGI by its effect on the economy: How does productivity increase?

After giving his definitions of both, he offers us this:

So, the two distinct questions we end up with are:

• Will there exist by Dec 31, 2030, an AI that is able to do every cognitive digital task equivalently or better than the best human, in an equivalent or shorter time, and for an equivalent or cheaper cost?
• Will by Dec 31, 2030, the U.S. have seen year-on-year full-year GDP growth rate of 19% or higher?
• Overall, what I would argue is the most important

Superforecasting technique is to look at the historical base rates and weigh these heavily in the analysis. At a high level, this means looking to the past for answers about the future and assuming that the past either repeats itself or at least rhymes, as Mark Twain might have said. Not assuming that the future will be radically different than the past is something that has served forecasters well for a long time. However, there are some key nuances that must be kept in mind and important choices to make when establishing the base rate.

After explaining something of his reasoning process, he offers us this:

These two questions have separate base rates and will likely therefore resolve differently. My forecast, with 60% and 90% likelihood, respectively, is that the first one will resolve in the affirmative and the second one in the negative. I.e., by 2030, it seems that we might have “AGI” but not “TAI”.

Here AGI is his input definition and TAI (Transformative AI, “powerful enough that the economy changes significantly”). So, he things it is fairly likely that we’ll have something that looks and smells like AGI by 2130, but it is unlikely to have much effect on the economy.

Here's my response:

I've got intellectual stakes in this game. On "AGI," "ASI," "PASTA," my basic impulse is "a pox on all your houses." But if we must use a term, I pick "AGI." My own prediction, with fairly high confidence (say >75%), is that we'll have substantial understanding of how LLMs work internally before AGI is reached. I make no prediction on when either of those will be achieved. I do think it possible that we'll achieve such understanding without ever reaching AGI, or that AGI will come much later than such understanding.

What do I mean by "substantial understanding"? Tricky question. However that understanding is expressed (e.g. a formal mathematical theory, engineering standards) it will allow us to explicitly engineer systems to meet specified performance objectives given specified operational capacities and input data for training.

Note that in arriving at the >75% figure I didn't undertake any explicit process. That number is pure spitballing.

One could reasonably say that my comment is not responsive to the argument he made. Why? What it comes down to is that Murray seems to think that the causal processing leading to AGI is homogenous, my term, while I don’t think it is.

To see what that’s about I’m afraid you’re going to have to read a dialog I had with Claude 3.7 about the nature of superforcasting. As always in these conversations, my remarks are in bold face while Claude’s responses are in plain face. Note that some headings in Claude's response will be in bold face, but that should be obvious from the context.

* * * * *

The Nature of Superforcasting

I understand that there is a formal discipline of forecasting associated with Philip E. Tetlock and the book Superforecasting: The Art and Science of Prediction. I wonder where it has been most successful, what kinds of phenomena? For example, we aren’t going to use it to predict the locations of the planets. Why not? Because that’s a well understood physical system where we can calculate planetary motion. Similarly, we’re not going to use it to predict the weather, and for much the same reason. Though in this case the system is complex, in a technical sense, and our ability to predict the weather is limited by the amount of data we have for estimating an initial state and the amount of computing power we have to run the prediction. I would also say that the causal processes in those systems are homogeneous in the sense that there is only one kind of causal process involved. OTOH, its not at all obvious to me that the stock market is homogeneous in that sense. I suppose the causal process is one involving the collective judgements of people trading in the markets. Those people have all kinds of reasons for making trades based on a wide variety of data available to them, including personal superstition. So, where has superforcasting been most useful?

I'll explore where superforecasting has been most successful based on the distinctions you've drawn between different types of predictive systems.

Superforecasting, as developed by Philip Tetlock and colleagues, has been most successful in domains with specific characteristics that fall between the deterministic physical systems you mentioned (planetary motion, weather) and completely chaotic or random systems.

The most successful applications of superforecasting tend to be in areas with these features:

1. Geopolitical events and international relations: Superforecasting has shown particular success in predicting political stability, leadership changes, and diplomatic conflicts. The Good Judgment Project (GJP) demonstrated superior accuracy in forecasting geopolitical events compared to intelligence analysts with access to classified information.
2. Economic trends and indicators: While not perfectly predicting specific stock movements (as you noted), superforecasters have demonstrated skill in predicting broader economic indicators like GDP growth, inflation rates, and unemployment figures.
3. Public health outcomes: The COVID-19 pandemic saw superforecasting techniques applied to predict case numbers, hospitalization rates, and vaccine development timelines with varying degrees of success.

What makes these domains suitable for superforecasting approaches?

• They involve complex causal processes that combine both systematic patterns and human judgment/behavior
• They have enough stability that expertise and historical patterns matter, but enough uncertainty that probabilistic thinking is valuable
• They involve phenomena where updating forecasts based on new information improves accuracy
• They benefit from aggregating diverse perspectives and avoiding cognitive biases

Superforecasting is less useful in domains where:

• Physical laws completely determine outcomes (planetary motion)
• Pure randomness dominates (lottery numbers)
• The time horizon is extremely long
• There's complete information asymmetry

The most successful superforecasting applications typically involve what Tetlock calls "Goldilocks problems" - not too hard, not too easy, where careful thinking and information aggregation can outperform both simple models and expert intuition. Would you like me to elaborate on any particular application domain or on the specific forecasting techniques that have proven most effective?

Supercasting the future of AI

In your judgment, is predicting the future of AI a plausible domain for superforcasting? Why or why not? Perhaps you could imagine a range of answers to that question, with reasoning appropriate to each answer within the range.

I think predicting the future of AI is a mixed case for superforecasting, with some aspects being amenable to this approach and others being more challenging. Let me explore a range of possible answers with reasoning for each:

Position 1: AI is well-suited for superforecasting techniques

This position holds that AI development has enough structure and pattern to make superforecasting valuable. Supporting reasons include:

• AI development follows somewhat predictable research and deployment cycles
• There's a large community of experts with diverse perspectives who could contribute forecasts
• Short to medium-term developments (1-3 years) often build incrementally on existing capabilities
• Many AI milestones involve measurable benchmarks (e.g., performance on specific tasks)
• Historical patterns of technology adoption can inform diffusion predictions

Position 2: AI is moderately suited for superforecasting with significant limitations

This middle position acknowledges both strengths and weaknesses:

• Near-term technical developments may be forecastable (e.g., when certain performance thresholds will be reached)
• Commercial adoption patterns might follow historically similar technologies
• However, discontinuous breakthroughs ("AI winters" and sudden advances) introduce unpredictability
• The field involves interacting technical, economic, social, and regulatory factors
• The most important questions often involve complex social impacts rather than just technical capabilities

Position 3: AI is poorly suited for superforecasting

This position argues that AI's future is fundamentally too uncertain:

• Revolutionary breakthroughs can occur unexpectedly, creating discontinuities that break trend-based forecasting
• The interplay between AI capabilities, human adaptation, and regulatory responses creates too many feedback loops
• Long-term impacts involve deeply transformative possibilities that lack historical analogues
• The most important questions about AI's future may be precisely those where we have the least precedent
• The field suffers from both hype cycles and secrecy that distort information needed for accurate forecasting

My judgment is closest to Position 2. I believe superforecasting can be valuable for specific, well-defined questions about AI with moderate time horizons (1-5 years), particularly regarding technical capabilities, commercial adoption patterns, and regulatory developments. However, the farther out we look and the more transformative the potential impacts, the less reliable superforecasting becomes for AI.

We've got liftoff...on a rebuilt and highly modified McLaren P1.

I've been following this saga for two years or so. Tavarish bought an almost new Mclaren P1 that had been submerged in a flood. So he got it, cheap. Could he make it run? That's what he's been doing for the last two years. In between doing this that and the other and living his life, he and his buddies have been working on this car. In the process they've decided to do more than bring it back to life, like brand new. No, that's not enough. They've decided to modify it so that it's better than new.

Anyhow, this is the first time they try to start the engine. And, it works! (c. 40:00) Look at the smoke streaming out! There's a bit more work to do. Just tinkering. Basically, the corner has been turned. Hear the engine roar. (c. 47:14) Joy breaks out (c. 50:29).

What will happen next? I don't know. Note that this video is four months old, so whatever is next, it's been happening for a few months now.

A ray of hope

The implosion of American culture and society: No way back?

Over at Linkedin futurist Bryan Alexander linked to this article: FBI raids home of prominent computer scientist who has gone incommunicado. Here’s how that article opens:

A prominent computer scientist who has spent 20 years publishing academic papers on cryptography, privacy, and cybersecurity has gone incommunicado, had his professor profile, email account, and phone number removed by his employer, Indiana University, and had his homes raided by the FBI. No one knows why.

Xiaofeng Wang has a long list of prestigious titles. He was the associate dean for research at Indiana University's Luddy School of Informatics, Computing and Engineering, a fellow at the Institute of Electrical and Electronics Engineers and the American Association for the Advancement of Science, and a tenured professor at Indiana University at Bloomington. According to his employer, he has served as principal investigator on research projects totaling nearly $23 million over his 21 years there.

Later in the article:

“None of this is in any way normal,” Matthew Green, a professor specializing in cryptography at Johns Hopkins University, wrote on Mastodon. He continued: “Has anyone been in contact? I hear he’s been missing for two weeks and his students can’t reach him. How does this not get noticed for two weeks???”

Here's the comment I made to Bryan’s post:

Yikes! Super-creepy. As you know, Bryan, I am a ronin scholar, have been for decades. I failed to get tenure in ‘85 and was unable to secure an academic post. For a while, perhaps almost two decades, my guiding fantasy was that I'd be able to write my way back into a really nice academic gig. But as I grew older that became less and less plausible, though I was able to keep my intellectual productivity up. Still, sometime in the 2010s I dropped that fantasy. By that time woke was rising and the academy was looking less and less attractive.

But now, if faculty can be “disappeared”... The academy is becoming a nightmare. Wokism came from within. And while the Trumpist crackdown is ostensibly a reaction to wokism, I can't help but feel that it is also the cumulative effect of decades of anti-intellectualism in American culture. This nightmare is coming from outside the academy. But it reflects the dark underbelly of American culture. We're imploding.

What's ahead? The Democrats surely want to go back, back to the world of Biden and Obama and Clinton. They can't figure out how to do it. I suspect that there is no way back. If that's right, if there really is no way back, then what's ahead? What's the alternative to Trumpism?

I note as well, that we’re facing the deployment of AI in society. That will make it ever more difficult to “go back,” by which I mean restabilize on the world as it existed in, say, 1999 (late in the Clinton administration), but with more computers. Yes, there will be more computers, but they’re pushing, pulling in a different direction.

We’ve reached an inflection point, though “point” is not the right term. We in a region of social and cultural phase change. Just how we come out of it is not strongly determined, nor is the exit zone. We could churn around for years, even for decades, or we could work our way though, perhaps not in years, but, yes, I’d think in decades.