Adrian Boutel
aboutel.bsky.social
Adrian Boutel
@aboutel.bsky.social
Kiwi in Suisse Romande. Have practised law and taught philosophy. Wondering whether to do WSET 4.
German budget lunatics shocked to find out that money is fungible.

on.ft.com/4qUndnM
Merz accused of using debt bazooka for welfare and tax cuts instead of investment
Two prominent institutes and the Bundesbank sound alarm over use of new borrowings
on.ft.com
November 11, 2025 at 6:55 AM
More delay for the Wings of Winter.
November 9, 2025 at 9:50 AM
All that murder, and he won’t even be a trillionaire. If those things vest I’ll eat a Tesla.
The world’s first trillionaire initiated a move that has left more than half a million people dead, most of whom are children.
November 7, 2025 at 9:27 AM
See also: managed exchange rates are libertarianism, socialism is when you let people exchange things at market prices
November 6, 2025 at 6:43 AM
“Forrester's analysis found that using AI for financially driven layoffs can backfire: 55 percent of employers regret laying off workers because of AI. More people in charge of AI investment expect it to increase headcount (57 percent) than to decrease it (15 percent) over the next year.”
AI layoffs to backfire: Half quietly rehired at lower pay
: Bosses banking on automation? 55% will regret those job cuts
www.theregister.com
November 5, 2025 at 12:28 PM
Caught the extra innings again this morning. Apparently the rest of the game was exciting as well?
November 2, 2025 at 7:11 AM
Holy crap

“They coldbloodedly killed everyone they found inside the Saudi hospital, including patients, their companions, and anyone else present in the wards,” a spokesperson for the Sudan Doctors Network said, adding that the RSF had transformed hospitals “into human slaughterhouses”.
FT
The latest UK and international business, finance, economic and political news, comment and analysis from the Financial Times optimised for your device on app.ft.com.
app.ft.com
October 31, 2025 at 6:12 AM
I thought this was just an abstract paradox.
October 30, 2025 at 11:43 AM
Pas de subventions pour les commerces de la #ruedecarouge qui font ceci :
October 30, 2025 at 7:47 AM
Reposted by Adrian Boutel
There are no words for Sudan-- the "blood bath" conclusion marks the complete failure of the international community. This is a horrific situation & like so many other wars, I doubt there will be justice for these survivors & their families.
Sudan siege ends in bloodbath despite pleas for mercy
Evidence emerges of atrocities committed by the paramilitary RSF after it seized control of El Fasher
www.ft.com
October 29, 2025 at 5:41 PM
This is right, of course, but the results do offer a corrective to anyone who thinks (or would like customers to think) that humanlike reasoning somehow emerges from all the associative inference. In particular, “chain of thought” promoting makes no exception.
Psychologists running empirical studies to rediscover engineering design choices is such a strange genre of papers. By all means, run studies on LLM judgments -- but what else than lexical co-occurence and statistical priors would they be based on??
Evidence that even when LLMs produce similar results to humans, they “rely on lexical associations and statistical priors rather than contextual reasoning or normative criteria. We term this divergence epistemia: the illusion of knowledge emerging when surface plausibility replaces verification”
October 27, 2025 at 5:48 AM
Reposted by Adrian Boutel
in the first part of Luttwak's Strategy: The Logic of War and Peace, he challenges armchair generals to organize a moderately large trip to the beach. After walking through all of the ways it goes wrong, he then drops "now imagine people are shooting at you"
October 26, 2025 at 1:14 PM
I’m going to charitably assume this is all a Saint-Ronald-would-never gotcha, and not people getting negatively polarised by Trump’s (absurdly high and chaotic) tariffs into stanning neoliberal globalisation.
Here’s the original clip of Ronald Reagan from April 25, 1987, where he delivered a complete and total rebuke against tariffs. Trump's calling Reagan’s words here “FAKE” and “fraudulent.” They’re 100% real. And the original clip is actually far worse for Trump, as much is left out of the ad:
October 24, 2025 at 6:32 AM
Only thing missing from this is a Greens surge. Otherwise, (4th) Tory implosion, (3rd) you can’t beat the right by tacking right, (2nd) racists losing, (1) I’m a cel-tic nation, get me out of here.
October 24, 2025 at 6:14 AM
Way to grab disaffected US scientists, Kier love.

www.ft.com/content/7144...
October 21, 2025 at 6:17 AM
Fortunately it won’t happen! I don’t know if the bubble will burst this year, or next, but by 2035 these things will be vast and trunkless legs of stone.
“By 2035, data centers globally are projected to use about as much electricity as India…according to the International Energy Agency. A single data center can also use more than 500,000 gallons of water a day, nearly as much as an Olympic-size swimming pool.”

🎁:
www.nytimes.com/2025/10/20/t...
From Mexico to Ireland, Fury Mounts Over a Global A.I. Frenzy
www.nytimes.com
October 20, 2025 at 12:47 PM
Gandria, lake Lugano
October 19, 2025 at 10:04 AM
No mention in this of *why* central banks had to engage in massive, asset-bubble-inflating bond purchases: because interest rates were at zero and governments were responding to recession with austerity instead of stimulus.

This, too, is on Osborne’s head.

on.ft.com/4mRUnBm
The populist shadow hanging over central banks and QE
The massive bond purchases are being unwound at a time of fierce political criticism of the monetary authorities. The controversy could limit options in future crises
on.ft.com
October 8, 2025 at 7:21 AM
“But without all the excitement around AI, the US economy might be stalling out, given the multiple threats.”

Only might? What’s the other bet? Tariffs and labour shortages might be good?

on.ft.com/4pTQ3US
America is now one big bet on AI
It’s seen as the magic fix for every threat to the US economy
on.ft.com
October 6, 2025 at 5:01 AM
Reposted by Adrian Boutel
A common confusion I'm seeing is people mixing levels of analysis wrt neural nets: we understand the implementation level well and the algorithmic level somewhat but not the computational level of "how does it internally compute things."
October 5, 2025 at 3:55 PM
People will always make all the errors, some will be too enthusiastic and some too sceptical of new tech. But in the current situation, of planet-gobbling construction, circular investment schemes and 9-figure salaries, is there one error we should be more worried about?
October 5, 2025 at 3:00 PM
‘“Knowing exactly how internal combustion engines work” is an obvious superset of “describe the trajectory of a particular butane molecule”.’

Why that’s false is a surprisingly deep metaphysical question, but if it’s not false then no one understands how *anything* works. #panpsychism
“knowing exactly how llms work” is an obvious superset of “describe how one small component of them works”
October 5, 2025 at 8:27 AM
Upper Cam, Darwin and Queens’, the morning after
September 30, 2025 at 1:20 PM
Venue looking good
September 27, 2025 at 10:35 AM