Newsletter
Evolution is overrated and other things
April 18, 2025
Hello,
Here’s everything since my last little missive to you:
Excerpt: People love a good evolutionary narrative. I wouldn’t be able to count the number of times I’ve heard “back in our evolutionary past…”. Somewhere along the line, evolutionary theories went from a useful way to fix psychological theories, to a generator of some of the most superficially idiotic. And I think, reading between the lines, we can find a new use for them. But first, let me convince you that evolutionary narratives aren’t usually worth very much.
Main idea: Without time-travel, evolutionary narratives can only identify theories that don’t make sense (like death drives). It can’t tell you what theories do make sense, because you can make many to explain the same thing. All they do is let you see what people wish the world was like.
Welcome to the Chaoscene:
The climate crisis is here. In order to thrive in these dangerous and precarious times, we must build resilient communities
A comment on the value of community in global instability. They point out:
Our society’s individualism, largely driven by technological advances and the illusion of endless progress, will no longer be sustainable – will not be able to last – in a world where anthropogenic climate change is a constant and growing threat.
But I think we recognise that. I think, like the 60’s, we are moving towards cultic business models again, because we are sensing what is going to be required of us.
–
AI tutoring works:
students working on mathematics with tutors randomly assigned to have access to Tutor CoPilot are 4 percentage points (p.p.) more likely to master topics (p<0.01). Notably, students of lower-rated tutors experienced the greatest benefit, improving mastery by 9 p.p. relative to the control group. We find that Tutor CoPilot costs only $20 per-tutor annually, based on the tutors’ usage during the study. We analyze 550,000+ messages using classifiers to identify pedagogical strategies, and find that tutors with access to Tutor CoPilot are more likely to use strategies that foster student understanding (e.g., asking guiding questions) and less likely to give away the answer to the student, aligning with high-quality teaching practices. Tutor interviews qualitatively highlight how Tutor CoPilot’s guidance helps them to respond to student needs, though tutors flag common issues in Tutor CoPilot, such as generating suggestions that are not grade-level appropriate
So a 4-9% improvement in learning outcomes, at the session level. But mostly it highlights that AI is helping bad tutors become as good as seasoned ones, which is more exciting.
–
Psychedelics demolish assumptions:
With their uncanny ability to disrupt our ordinary modes of perception and thinking, psychedelics have long been associated with feelings of insight … One of the most powerful cumulative effects of these disruptions is what I call ‘epistemic loosening’. Epistemic loosening is, in its essence, a quasi-Socratic, temporary destabilisation of deeply held beliefs and assumptions
I talk about the ‘language problem’ all the time. So here is a new way of talking about this kind of intuitive insight, and as a bonus, how psychedelics interact with it. And here are some old articles of mine that talk about the same thing from other angles.
–
Shortage predictions from tariffs. I don’t actually care much about this, because I can never predict the likelihood of this stuff. But I was curious if AI could. So here’s the tweet prediction:
In ~3-6 months we are going to start having shortages of transformers, pumps, air conditioners, and other complex goods which are made (or their critical components) in China.
All because of the tariffs. And here’s what GPTo3 reckons will happen to markets.
– China’s past retaliatory measures have been targeted (rare earths, Micron ban) rather than blanket export stops on mid‑value, high‑volume goods that employ hundreds of thousands of Chinese workers. – Domestic capacity (ABB, Eaton/Cooper, SPX, WEG plants in the Carolinas, TX, MO) can cover roughly one‑half of normal U.S. demand for distribution transformers and is being expanded with federal grants; not great, but enough to avoid blackouts. – HVAC/Pump supply is more diversified—Thailand, Mexico, and Korea already account for >45 % of U.S. room‑A/C imports. – Timeframe: new tariffs can start in 30 days after notice, but actual shortages would show up after 4‑6 months if inventories are thin. Utilities usually carry 6–12‑month spares on critical units.
Probability estimate: 10‑20 % chance of noticeable, nationwide shortages inside six months; <5 % chance of a “collapse of basic services.” More likely outcome is higher prices and selective delays, not empty shelves.
Though, note it’s only trained to June last year, so adjust your estimates there because it’s worried about an election year. And the market activity:
- Likely near‑term losers if severe shortages materialise
- HVAC & building‑systems OEMs: Carrier (CARR), Trane (TT), Johnson Controls (JCI), Lennox (LII). Margin squeeze if they must air‑freight parts or halt some lines.
- Electric‑equipment distributors: Wesco (WCC), Rexel (RXEEY) rely on steady transformer flow.
- Retailers with big seasonal A/C sales: Home Depot (HD), Lowe’s (LOW), Best Buy (BBY).
- Grid‑mod/utility cap‑ex plays: Quanta Services (PWR), AEP, Duke (DUK) could see project delays.
- Small‑cap pump makers with China content: Franklin Electric (FELE), Graco (GGG), AO Smith (AOS).
- Import‑dependent logistics names: Matson (MATX), Expeditors (EXPD) get volume hit if trade slumps.
Possible winners / relative beneficiaries
- U.S./EU transformer makers with domestic plants: SPX Technologies (SPXC), Eaton (ETN), Hubbell (HUBB), ABB (ABB), Siemens (SIEGY). Could receive rush orders and favorable pricing.
- Latin‑American & ASEAN HVAC suppliers: Daikin (Japan, 6367.T), Midea‑listed appliance subsidiaries, Intercable (Mexico, private) → indirectly positive for local industrial REITs like Prologis (PLD) with Mexico exposure.
- North‑American rare‑earth or magnet plays that displace Chinese supply: MP Materials (MP)— though MP still sends concentrate to China for final processing, so benefits are capped until the Texas plant ramps.
- Domestic steel/copper producers if utilities scramble for conductor material: Nucor (NUE), Freeport‑McMoRan (FCX).
- Tariff‑proof “Buy America” ETFs: PAVE (infrastructure), XLI (industrials) could outperform broad market.
Market behaviour to watch
- Spot transformer lead‑time (published weekly by E‑Source). If it jumps from the current 40‑50 weeks back to 70‑90 weeks—2019 tariff‑war levels—utilities may guide cap‑ex delays.
- CARR, JCI input‑cost commentary on earnings calls (they flag compressor and ECM motor sourcing).
- Freight indices (Freightos Asia‑US East Coast) for a sudden collapse in volumes.
- Filings: Any Section 301 list that explicitly adds transformers or HVAC units is a red flag.
So price inflation and episodic delays, giving domestic manufacturers bargaining power. Let’s see. I’ll keep my eye on SPX, myself. I won’t invest in it though—historically that’s a surefire way to tank the price of something.
–
We’re wrong about ADHD. Use a paywall buster:
some scientists have begun to argue that the traditional conception of A.D.H.D. as an unchanging, essential fact about you — something you simply have or don’t have, something wired deep in your brain — is both inaccurate and unhelpful. According to Sonuga-Barke, the British researcher, the traditional notion that there is a natural category of “people with A.D.H.D.” that clinicians can objectively measure and define “just doesn’t seem to be the case.”
We’ve known for a while that diagnosis is a bit helter-skelter, and self-diagnosing online is making it much harder to tell true cases from people who are learning how to meet diagnosable criteria (because ADHD is now trendy). But this is new and exciting—that ADHD might not be permanent. It might be environmental, and the lasting effects come from the effects of stimulants in making people feel a connection to their work. For people who are truly worried about ADHD, this is interesting to explore.
–
I hope you found something interesting.
You can find links to all my previous missives here.
Warm regards,