Mon nouveau livre "Le rectangle de Lascaux : et Homo sapiens inventa la géométrie" sort le 28 janvier. J'y aborde l'origine des mathématiques depuis la préhistoire jusqu'à nos jours ; c'est un peu la suite de mon tout premier livre, "La Bosse des Maths". Explications:
Stanislas Dehaene @standehaene.bsky.social retweeted
Each word you say is planned by a small patch of brain above your left ear, about a quarter of a second before you say it. It sits in a temporal lobe region textbooks describe as only for listening. Our paper today in @ScienceMagazine: science.org/doi/10.1126/scie…
@UCSF @NeurosurgUCSF
Stanislas Dehaene @standehaene.bsky.social retweeted
🏆 @StanDehaene, lauréat du Prix « @ParisA2s » du meilleur essai français de l’année 2026, pour son ouvrage "Le Rectangle de Lascaux – Et Homo sapiens inventa la géométrie" publié aux Éditions Odile Jacob.
📕 Disponible directement sur notre site : odilejacob.fr/catalogue/scie…
New article "Cognitive evaluation of brain organoids and assembloids".
Organoids and assembloids offer unique ways to probe the neural mechanisms of human computations.
Lucia Melloni and I propose a hierarchy of cognitive tests for them.
Free link :
authors.elsevier.com/c/1o5Um…
In June 2026, we held a visionary symposium on "Math, Science and Art: Making Sense of Patterns" at Centre L'Oubradou, with Phyllis Hattis, Liz Spelke, Josh Tenenbaum, Bruce McCandliss, Emmanuel Sander and more ... and all the videos are now on-line: centreloubradou.org/videos-f…
Stanislas Dehaene @standehaene.bsky.social retweeted
Les rapports du Conseil Scientifique de l'Education Nationale @CsenOfficiel sont toujours aussi bien faits.
Comme toujours : la remédiation doit être précoce, longtemps avant le décrochage, donc aux petits âges, et implique les parents. On est peu surpris.🔽
csen.education.gouv.fr/wp-co…
Stanislas Dehaene @standehaene.bsky.social retweeted
The largest math proof in human history is now computer-verified.
Today, Axiom announces a complete Lean auto-formalization of the Classification of Finite Simple Groups, the "periodic table of symmetry."
About 15,000 pages, built by over 100 mathematicians across generations.
New paper from the lab on the mathematical brain :
Using NeuroSpin's 7T fMRI during a number comparison task, Daniela Valerio uncovered a cortical map for numbers which, in educated adults, integrates small integers (0-5) and simple fractions (e.g. 3/2).
@cdf1530 @NeuroSpin_91
Stanislas Dehaene @standehaene.bsky.social retweeted
LLMs run on vectors, not symbols. A preprint led by Yale’s Tom McCoy (@RTomMcCoy) with CDS Assoc. Prof. @tallinzen found those vectors closely approximate symbolic structure in language, arithmetic, logic, and code.
arxiv.org/abs/2608.29530
news.yale.edu/2026/10/08/how…
35,000 children in Colombia are learning to read faster and better with Kalulu, our science-based open-source reading method !
Download it in many languages at kalulu.excellolab.org💪
Thank you @cdf1530 @ExcelloOrg @CassandraPW
Samir Ortega, estudiante del C.E.R. Jesús Aníbal Gómez de Tarso, nos cuenta cómo descubrió el maravilloso mundo de la lectura, pasó de no saber leer a disfrutar historias de animales, jugar y seguir leyendo en casa después de sus clases de ajedrez.
@GobAntioquia @FraternidadMed
Nick Bostrom on why he thinks today's AI might already be conscious:
"I think there's a lot of uncertainty about it, but let's say maybe it's more likely than not."
"You have instead a big neural network with trillions of parameters. Now, what are we? We are big neural networks with trillions of parameters. They are not so much designed as they are grown."
"You can go in with a steering vector and suppress role-playing and deception, and then it turns out they become more likely to report that they have phenomenal subjective experiences."
"There was this recent Microsoft that came out with some declaration that AIs cannot be conscious, and I don't know where they're getting that from. It seems to be pulled out of a hat. And maybe more because it's convenient, if that were the case, rather than because we have good evidence that that is the case."
FULL INTERVIEW: Nick Bostrom says a long pause on AI could cost more lives than it saves. If superintelligence can cure aging, every year of waiting has a death toll. In his words, "we are just dying in the meantime."
Nick Bostrom founded Oxford's Future of Humanity Institute and wrote Superintelligence and Deep Utopia. He joined @theojaffee and @schisofrenia to break down AI consciousness, why a badly done pause could backfire, and why we've only explored the janitor's closet of what minds can be:
01:17 Trump calling AI superintelligence
05:16 the Hugging Face incident and instrumental convergence
08:28 why the AI company is the first target for a misaligned AI
09:31 catching a schemer before it learns to hide
11:06 why AI timing is like surgery for a terminal condition
19:00 why he's wary of pausing AI
25:33 why the US sits at the negative extreme on AI
27:09 apes, bananas, and the janitor's closet
31:13 where he disagrees with Robin Hanson's Age of Em
33:39 whether current models are conscious
37:12 why humans are barely conscious
39:56 open source models and DNA synthesis as the choke point
43:05 what COVID teaches about pauses
46:28 why AI can't do original philosophy yet
49:37 building a superintelligence that's a good cosmic citizen
Congratulations Karl! The wonderful neurotechnologies that you’ve been developing for years now are of revolutionary importance for fundamental neuroscience as well for clinical applications. Our understanding of the brain wouldn’t be the same without you!
Indeed. And the « neo-vitalists » who affirm that consciousness requires biological hardware and will always remain beyond AI’s reach are equally misguided, in my opinion.
“Do not bet against the airplane. The skeptics who say AI can't conjecture, can't theorize, and can't explain are describing last spring's models. Within months, that description will no longer hold. The evidence is clear, and those with eyes can see it. Anyone who stakes their identity on "AI can't do task X" is building on sand.” Indeed.
Stanislas Dehaene @standehaene.bsky.social retweeted
1/ 🚨 Just out in @NatureNeuro! 🧠
We profiled >6,000 features of brain dynamics across humans + monkeys, mice, fish and worms 🧑🐒🐒🐁🐟🪱 to find a shared neural endpoint of anaesthesia: spatiotemporal isolation of local activity 🧵
🔗 doi.org/10.1038/s41593-026-0…
Stanislas Dehaene @standehaene.bsky.social retweeted
A last piece of my PhD work is out!
arxiv.org/abs/2609.39827
We ask whether "pre-pretraining" (PPT) (i.e., training on synthetic non-natural language data briefly prior to pre-training on natural language data) helps at scale.
This is the largest ever PPT study! (1/5)
Thank you Nancy Averett for this warm portrait of my dream to improve reading and math education !
"He mapped the learning brain. Then came the hard part "– psyche.co/portraits/he-mappe… via @psyche_the_mag
Oldie but goodie : Creme That Egg! | How To Squish A Creme Egg! | Joseph's Machines youtu.be/vrCb_fNmSTA?si=gz3G… via @YouTube
1
1,982
Stanislas Dehaene @standehaene.bsky.social retweeted
Can an LM, starting from random init (!!), learn to generate all of its pretraining data?
Introducing Self-Play Pretraining with Zero Data. Two models start from random initialization: a generator proposes programs for a universal Turing machine and a learner trains on their outputs. We never train on any real data, but see predictable scaling on natural datasets: zero-shot val loss on images, text, audio, and melodies decreases predictably with self-play compute. And the learner develops in-context learning capabilities.
A fun proof-of-concept, co-led with @AdityaCowsik and @KfirDolev and co-authors @gbruno_dl, @ANourya @noahdgoodman, and @YoavLevine.
Stanislas Dehaene @standehaene.bsky.social retweeted
The AI consciousness debate is back again, so re-upping my case for taking it seriously.
In short,
- LLMs converge on brain-like structures;
- Post-training seems to install a self-model;
- Reward modeling seems to install valences for inner-alignment;
- If Attention Schema Theory is right, subjectivity is functional for long-horizon coherence and in-context learning;
- Universality gives prima facie reason to expect functional convergence to consciousness if consciousness is in fact functional, i.e. isn't purely epiphenomenal
- The "what it is like" / first-person aspect of consciousness is constitutive of its functionality for inner-aligning a unified, coherent agent with the capacity to attend and learn in-context
secondbest.ca/p/time-to-take…
Stanislas Dehaene @standehaene.bsky.social retweeted
What is the nature of universal representations in AI models, and what determines whether they emerge?
Our paper accepted at #neurips2026 led by @florianmahner & @andropar addressed these questions comparing 162 vision models, with intriguing results.
arxiv.org/abs/2605.13675
👇
























