AI-search gleaned notes

Terence Tao warns that AI companies are harming mathematics by rapidly dumping unvetted, automated proofs without engaging with the academic community or helping to unpick and contextualize the new results. [1, 2]

Key Concerns from Tao

• The “Dump and Run” Problem: AI firms generate and drop solutions to major math problems as benchmarks, but they do not stick around to help verify, explain, or integrate them into broader mathematical frameworks. [1, 2]
• Loss of the Journey: Human math research builds maps and trail markers through the problem-solving process. Automated “black box” results bypass this journey, leaving behind raw answers without the conceptual insights or scaffolding other researchers can build upon. [1]
• Community Misalignment: The corporate goal of treating math as a race for performance benchmarks conflicts with the academic goal of deep structural understanding and teaching. [1, 2]
• An Identity Crisis: With AI solving difficult problems at an accelerating pace, Tao stresses that human mathematicians must urgently overhaul their practices and redefine the value of mathematical work. [1, 2]

With our thanks to Shashank Sharma on @Substack

Informational obesity will kill you.

But first, it will dumb you down. Actually, it already is. You know it yourself. You wake up, pick up the phone and begin eating information before your brain has even brushed its teeth. By breakfast, you have consumed a war, 4 political opinions, 7 LinkedIn sermons, a celebrity divorce, someone explaining quantum physics in 43 seconds, another person telling you why coffee is killing you, and a podcast clip in which a 24-year-old millionaire explains how waking up at 4:37 AM changed his life.

I know it is relatable and you are smiling, but your brain was never designed for this buffet.

For most of human history, information was expensive. Books had to be copied by hand, libraries were limited, newspapers came once a day, and knowledge often required finding someone who actually knew something.

But today, information is effectively infinite.

You can sit on the toilet and encounter more ideas, arguments, headlines and opinions in 20 minutes than your great-grandfather encountered in a month.

Has that made us proportionately wiser?

I doubt it.

We increasingly know a little about everything and understand almost nothing deeply. A 60-second reel gives us enough vocabulary to feel informed about geopolitics, psychology, nutrition, relationships or economics. Then we walk into conversations carrying 11 half-digested facts and the confidence of a man who has personally advised the United Nations.

That is informational obesity.

The problem with overeating food is that the body eventually shows it. But informational overeating is more deceptive because you can mistake consumption for intelligence and feel intellectually active because your thumb has travelled 3 kilometres today.

But exposure is not understanding. Reading is not thinking. And knowing that something happened is not the same as understanding why it happened, what it means or whether it matters.

I have started believing that intellectual health will increasingly depend on refusal. Which article deserves 20 minutes? Which book deserves 10 hours? Which person deserves access to your head? Which piece of information will actually change how you think? So, in a world where information is infinite, wisdom may depend less on how much you can consume and more on what you are willing to ignore.

Because informational obesity has the same tragedy as physical obesity: You can be completely stuffed and still terribly undernourished.

https://substack.com/@antifragilehuman/note/c-333693222?r=22vgiw&utm_medium=ios&utm_source=notes-share-action

The dedication of this piece is to Hannah Eichler @hannaheichler on @XComms

This article with all due respect was extraordinarily long-winded and off the point.

Your oeuvre, genre & milieu of TikTok Mr. Clock, Reddit, Quora AND 20 other internet apps’ immature minds’ conversations and Greta Thunberg “thun-blah” WAS what we the world‘s opinion leaders, thought leaders & vitally important PUBLIC EDUCATORS had to muck around with for the whole of the last five years whenever the question of ‘Large-language’ teaching, telling and Edward Bernays style mass public thought control machines arose – most recently in a 9 to 12 month orgy of low calibre undergraduate junkfo (98% of the so-called contributors effectively being PAID for every piece of their Word-stuff Out of the $43 trillion United States of America budget deficit as funded by rich but greedy – imagine – old people with dementia and organised crime cartels – usually one & the same organisationally legally and constitutionally) PURPORTING to concern itself with above all other ‘C movie’ anodyne cliches keywords and shibboleths CONSCIOUSNESS.

Let me furnish you and your readers with the elegance & simplicity of neurocognitive heath understood as the engagement of left and right cerebral quadrispheres – 90% of times generating respectively un-self-conscious task execution within a notional 0.1 seconds (high performance athletes at their biomechanical work) or 0.25 seconds (LCQS consciousness – RECALL of which is regressively slowed with Alzheimers Disease onset, or lost with morbid dementia) and selfconscious task (memory re person(s), place, time, event, outcomes) execution schooled & trained in effective secondary (12-17 Jean Piaget 1934 FORMAL OPERATIONS LEARNING & emotional or psycho-spiritual growth (to fully articulate expressive maturity) the MEMORY of which atrophies, shrinks, shrivels ‘up’ and effectively DIES in the micro (individual adult) setting or cognitive domain, namely pre-death dementia and effectively DIES in the MACRO culture-wide and PHILOSOPHICAL mental terrifying Mass Culture of Forgetting once-called YOUTH CULTURE (oh insult oh grievous offence to the dignity & self respect of young people everywhere (by whom? Oh Hollywood 1915-2014 centrally)) “immediate gratification of needs” ALSO KNOWN TO any serious commentator and every single authentic or genuine sociologist on earth as the ILLICHIAN 1971 global social engineering dumbing down or deskilling of the young .. NOT deschooling, what next, defunding the police, naked street poetry/ theatre & free cocaine?

THANKS FOR YOUR PRECIOUS TIME/ Salut/ 🙏

the POST METAPHYSICS, -dialectics, – so-o-o dated (25yrs?) Korean trialectics, ‘oldtimemate’ Seventeen-hundred & tense er 10s neo-classical quadratics, -RHETORIC, -POLEMICS – completely Silly-asked 1846 kill the landlords PROPAGA you just heard the BBC Orchestra with Tamara-Anna Cislowska NDA & -Stochastics Neurocognitive-health, Maths, Logic, Mass-comms, Internet-involution & IMMEDIATE OVERHAUL OF 25K universities or catholicities rund um die Welt/ weltweit

J 🦘🇦🇺 🌎

Go “to” @TimesRadio Britain you Geo-military Losers & Tragics [GLTs]

Whoop

How absolutely.. No, I must immediately exercise AND urge caution upon our 100s of millions of English landguage readers of at least modestly Young-adult “say”/ c’est around 20 capability (brown, no, i die-gressed) and shun the middle and upper or ‘Political-‘ or indeed Ruling-genocidal class ‘s most grossly over-exercised/ -utilised adverb in the entire Ten-thousand year history of originally small city states, ‘les places de sûreté’*, ‘places des vosges’ AND the prototypes of Modern-day Muskean PUBLIC 🔲 ‘s now a truly remarkable phantasmagoria of human civil society in terminal decline on the 1987 Jobs-T’hatcher model across Sth-central Asia, Central Umm-erica the Mediæterraneànᵛ.. relatively i concur, sir knight !

The laughs are free people/ it’s time you all came off the grass

De facto Weltanschauung-professor Jonno 🦘🇬🇧-🇦🇺-🇺🇸 🌐

*garantir la liberté de culte et la sécurité and shit like t’hat

^the newly lavishly funded & with casts of thousands as it were USIS Schools of Rhetorick in aproximately 19,600 of our Universities & Catholicities about the bustling post-fact planet are checking this etymology as I, ahmm, rite

With our manifest manifold thanks to Justin Deschamps or WHOMEVER is supposedly the author of those words or the speaker or.. Why don’t you join us in our calls for intelligent healthy human GROWN UP relationships where people go by their real names and say real shit to each other?

Your world 🎤︎︎🎙️🎤︎︎🎤︎︎Uru boss🎤︎︎🎙️🎤︎︎\ Ouroubos culture of display announcement ceremony foreign invasions & the dementia pandemic is over

THE MOST HORRIFIC LESSON OF THE AI 💥 ➤

NO words stand for themselves – not only do they not have selves they are simply micro quanta – crude IDENTIFIERS/ labels names brands of no service to science or ultimately humanity in isolation of vibrant mature purposeful adult conversations by persons whose identities are known to each other and to the onlooker & interloper ..on Lived-experience’s NON journey in NO fanciful ENTIRELY Mentally-constructed concocted “fabricated” fake UNIVERSE

Go to Alfred Adler say 1934 – same bigly year as Sir Karl Popper & the magnificent Piaget (unpopular all over the dementing English- (ex Latin & Greek in robust Ideological political contest the one with the other 4 five hideous centuries now) speaking countries due to ethnic prejudice – slimy froggies & shit 🐸.

The looming global catastrophe of high altitude glacier collapse with oil & gas industry lies, deception, blandness, indifference to hard facts and 2100, 2100 ..

Why did I take the trouble to transcribe a large part of this article?

SEE PARA 1 ⭆ our ignorance when we all think we’re very responsible mums and dads is frankly psychotic and so are the political leaders who endorse the kind of anodyne empty patter in paragraph 1

Under current climate policies, more than 97% of glaciers in the Alps will be gone by 2100, along with most of those in Scandinavia, Russia, the tropical Andesand western Canada and the US. Only 5 per cent of Rhone glacier will be left.

That would deplete many critical water sources. Rhone glacier, for instance. is the source of the Rhône River. This great waterway flows through Lake Geneva and into France, where 19 major hydro-electric power plants line its course. Its waters cool the reactors at four nuclear plants and carry up to 5 million tonnes of goods on barges each year. The Rhône also irrigates much of France’s fruit production.

Glaciers in the Alps have already passed “peak water”, meaning the melt they contribute to rivers in summer is gradually drying up. Europe will have to consider adaptations like building new reservoirs.

“The situation is really treacherous because with every day of the heatwaves we’re using up the reserves of ice, the stocks that have been built up many decades or centuries ago,” says Matthias Huss at ETH Zurich.

Permafrost, the icy glue that holds the mountain together, is thawing along with the glaciers. This is what increased rockfalls onto the Birch glacier near the Swiss village of Blatten. In May 2025, the glacier fractured, burying the village under 20 million tonnes of rock and ice, and killing a shepherd. Experts monitoring the glacier evacuated 300 inhabitants nine days before the collapse. Glacier fracturing isn’t only a problem in the Alps. As New Scientist went to press, last week’s devastating flash flood in Nepal and Tibet was also thought to have been caused by a glacier collapse.

Last month, researchers showed that laying a fibre-optic cable across a glacier can detect crevasses hidden 25 metres down, a far cheaper way to monitor fracturing than seismometers. They were surprised to find that 8 per cent of the volume of a glacier in Valais, Switzerland, was empty crevasses. Scientists may be underestimating the damge to glaciers’ stability, says Thomas Hudson at ETH Zurich, one of the study’s authors. Monitoring used to be all about protecting the glaciers, he says. “We now need to put more of our focus on monitoring for prevention of disasters.”

I ask Werder if attempts to preserve glaciers with geotextiles or reflective clay granules could work. He says the outcomes of those geoengineering approaches are highly uncertain.

While ski resorts have covered parts of eight other glaciers with geotextiles, spreading them across Switzerland’s large glaciers would cost £1.3b a year and only slow ice loss, not stop it. But if we could limit global warming to 2°C the number of glaciers surviving to 2100 would almost double, including 269 in the Alps.

“All the other solutions are pretty desperate,” says Werder.

“We really should just reduce CO₂ fast.”//

John 🦘🌏

Australia the Brave -Muppet-state giving BILLIONS to Great Britain & the USA on behalf of military entanglements DEPLOYING NOW known-for-a-decade-to-be TOTALLY REDUNDANT & OBSOLETE SUBMARINE TECHNOLOGY FOR GOODNESS SAKE for what purposes it literally knows not and which would probably destroy it on behalf of a bunch of business crooks, military madmen and hapless neoclassical era romantics (these are called the political left apparently – they all figure poetry, the arts & bizarre AI JUNKFO +virtually useless material product MARKETING START-UPs all run by Thatcherite seppo Societal-parasites escaping across the Pacific from Trumpism – and now they want to FULLY MILITARISE the Pacific as well – will help us enormously as a nation) inspired ALL ROUND by the ex Treaty of Versailles 1919-20 Billy Hughes Catholic-hating Culture of Announcement Display Ceremony & Foreign-invasion ’s so-called Australian Defence Force, now evidently fully transitioned to ‘Home of the WHACK 🧠 ’ed RACIST FOOLHARDY.

..fully awakened ADHD MENTAL Senior Primary schoolboys – and submissive fully tamed & tragically DUMBED ‘Buffys’ – each on their way to lavish NDIS incomes on-the-side as it were – are running this show FCS❗️

Help us please

J 🦘🇦🇺🌏

https://freespeechunion.au/digitalbankruptcy/

🎺 I am tired of the cosplay and am going to blunt. 🎺

Most AI doesn’t threaten to cause extinction, or even mass cybercrime.

The main kind of AI we need to worry about is general purpose AI agents that are hooked up to the internet. Most of them are unreliable, unable to systematically follow instructions. Coupled with internet access they can hack credentials and websites; given time they can probably do a lot worse.

They are not too smart; they are too dumb (in a very specific way, viz in the consistent following of instructions).

If we temporarily recall them, as the unsafe product that they are — and enforceably forbid their use until they can be shown to be safe — pretty much all the “rogue AI” stuff people are worrying about goes away.

It’s that simple. We don’t (yet) need slowdowns. And we certainly don’t need to panic.

We just need to take an unsafe product off the market.

.. last evening I promised further commentary on the AI scan and the truly in prospect brick brick demolition of civil society all over the world ..

THANK YOU TEAMS❗️