Thread of deep reasons behind simple facts: https://x.com/anderssandberg/status/2053757849918939364
Many “separate” theorems are the same structure seen in different clothes: Stokes unifies integration theorems; RG explains why Gaussians are universal; diagonal/fixed-point arguments unify major impossibility theorems; Noether turns symmetry into conservation; Legendre duality connects mechanics, thermodynamics, and optimization; exponentials are the Lie-theoretic bridge from infinitesimal addition to finite multiplication.
Many “separate” theorems are the same structure seen in different clothes: Stokes unifies integration theorems; RG explains why Gaussians are universal; diagonal/fixed-point arguments unify major impossibility theorems; Noether turns symmetry into conservation; Legendre duality connects mechanics, thermodynamics, and optimization; exponentials are the Lie-theoretic bridge from infinitesimal addition to finite multiplication.
🔥3🤡1
This media is not supported in your browser
VIEW IN TELEGRAM
Thinking Machines: People talk, listen, watch, think, and collaborate at the same time, in real time. We've designed an AI that works with people the same way.
Read more: https://thinkingmachines.ai/blog/interaction-models
Read more: https://thinkingmachines.ai/blog/interaction-models
🥱9🤔1🤡1
This media is not supported in your browser
VIEW IN TELEGRAM
Minister of Defense Mykhailo Fedorov met Palantir CEO Alex Karp in Kyiv to expand AI and defence tech cooperation. Ukraine says work with Palantir has created a detailed air attack analysis system, deployed AI tools for large intelligence datasets and integrated technology into deep strike planning. Brave1 Dataroom now gives developers real battlefield data to train AI models.
😭10👍6🤡4💩3❤2🔥1
Links for 2026-05-12
AI
1. Bloom filters in NNs: gradient descent found an algorithmic solution, not vague approximation. https://www.lesswrong.com/posts/buxBdp8NtHGgBwabv/neural-networks-learn-bloom-filters
2. Neural Computer: a new machine form. https://metauto.ai/neuralcomputer/
3. Recursive Agent Optimization (RAO). https://apga.github.io/RAO/
4. LLMs Improving LLMs: agentic test-time scaling. https://arxiv.org/abs/2605.08083
5. Self-Improving Pretraining: post-trained models pretrain better models. https://arxiv.org/abs/2601.21343
6. 7B Conductor hits SOTA on GPQA-Diamond and LiveCodeBench by orchestrating LLMs. https://arxiv.org/abs/2512.04388
7. Recursion Is The Next Scaling Law In AI. https://www.youtube.com/watch?v=DGtUUMNYLcc
8. Recursive Multi-Agent Systems. https://arxiv.org/abs/2604.25917
9. Language Models Can Autonomously Hack and Self-Replicate. https://www.lesswrong.com/posts/JuoDNYDG8CgiQaCcz/linkpost-language-models-can-autonomously-hack-and-self
10. Automating AI research and explosive growth: innovation-network feedback loops. https://www.nber.org/papers/w35155
11. Interaction Models: scalable human-AI collaboration. https://thinkingmachines.ai/blog/interaction-models/
12. DynamicVLA: VLA model for dynamic object manipulation. https://www.infinitescript.com/project/dynamic-vla/
13. AI-guided discovery of atypical protein assemblies. https://www.tsl.ac.uk/publications/167424
14. Perceptron Mk1: video and embodied reasoning. https://www.perceptron.inc/blog/introducing-perceptron-mk1
15. Notes from inside China’s AI labs. https://www.interconnects.ai/p/notes-from-inside-chinas-ai-labs
16. Nvidia’s Jim Fan: 95% certain robotics tech tree unlocks by 2040. https://www.youtube.com/watch?v=3Y8aq_ofEVs
17. Is ProgramBench Impossible? https://www.lesswrong.com/posts/3pdyxFi6JS389nptu/is-programbench-impossible
18. GPT-5.5 & Opus 4.7 on ARC-AGI-3. https://arcprize.org/blog/arc-agi-3-gpt-5-5-opus-4-7-analysis
19. Sakana AI simulator: five “neural cellular automata” species learn, attack, defend, and compete on a 2D grid. https://www.youtube.com/watch?v=QzZ4VwDHAT4
IQ
1. IQ With Conscience. https://www.betonit.ai/p/iq_with_consciehtml
2. German study: higher-IQ people judge others’ intelligence better. https://www.sciencedirect.com/science/article/pii/S0160289625000972
3. Scott Wu interview: Cognition AI co-founder, elite US programmer, card-trick/mental-math prodigy. https://colossus.com/article/scott-wu-tapes-cognition/
4. Magnus Carlsen on John Nunn: too much understanding and curiosity can distract from chess. https://infoproc.blogspot.com/2013/11/sven-magnus-en-carlsen.html
5. Chinese-born AI researchers in Silicon Valley: $500k salaries, $50m earners, $200m poaching rumours. https://restofworld.org/2026/chinese-ai-researchers-silicon-valley/
Neuro(tech)
1. Printed neurons communicate with brain cells. https://news.northwestern.edu/stories/2026/4/printed-neurons-communicate-with-living-brain-cells
2. Brain-scanning gadgets are coming. https://www.wired.com/story/get-ready-for-more-brain-scanning-consumer-gadgets/ [no paywall: https://archive.is/k3YSe]
3. How neurons sense gut bacteria. https://news.mit.edu/2026/how-neurons-sense-gut-bacteria-0430
4. New theory of imagination in the brain. https://singularityhub.com/2026/04/30/how-does-imagination-really-work-in-the-brain-new-theory-upends-what-we-knew/
Miscellaneous
1. Prediction markets: why almost everyone loses except a few sharks. https://www.wsj.com/finance/investing/polymarket-kalshi-betting-profits-prediction-markets-eb23ac11 [no paywall: https://archive.is/EbW0e]
2. Nature genomics reframes the “barbarian invasion” as slow local transformation. 258 southern German ancient genomes show northern-ancestry groups beside Roman provincials before 476 AD, mostly endogamous. After imperial collapse, intermarriage accelerated; within ~150 years the population was close to modern Central Europeans: not tribes overrunning Rome, but kin groups, labourers, families, and communities reshaping society. https://www.nature.com/articles/s41586-026-10437-3
AI
1. Bloom filters in NNs: gradient descent found an algorithmic solution, not vague approximation. https://www.lesswrong.com/posts/buxBdp8NtHGgBwabv/neural-networks-learn-bloom-filters
2. Neural Computer: a new machine form. https://metauto.ai/neuralcomputer/
3. Recursive Agent Optimization (RAO). https://apga.github.io/RAO/
4. LLMs Improving LLMs: agentic test-time scaling. https://arxiv.org/abs/2605.08083
5. Self-Improving Pretraining: post-trained models pretrain better models. https://arxiv.org/abs/2601.21343
6. 7B Conductor hits SOTA on GPQA-Diamond and LiveCodeBench by orchestrating LLMs. https://arxiv.org/abs/2512.04388
7. Recursion Is The Next Scaling Law In AI. https://www.youtube.com/watch?v=DGtUUMNYLcc
8. Recursive Multi-Agent Systems. https://arxiv.org/abs/2604.25917
9. Language Models Can Autonomously Hack and Self-Replicate. https://www.lesswrong.com/posts/JuoDNYDG8CgiQaCcz/linkpost-language-models-can-autonomously-hack-and-self
10. Automating AI research and explosive growth: innovation-network feedback loops. https://www.nber.org/papers/w35155
11. Interaction Models: scalable human-AI collaboration. https://thinkingmachines.ai/blog/interaction-models/
12. DynamicVLA: VLA model for dynamic object manipulation. https://www.infinitescript.com/project/dynamic-vla/
13. AI-guided discovery of atypical protein assemblies. https://www.tsl.ac.uk/publications/167424
14. Perceptron Mk1: video and embodied reasoning. https://www.perceptron.inc/blog/introducing-perceptron-mk1
15. Notes from inside China’s AI labs. https://www.interconnects.ai/p/notes-from-inside-chinas-ai-labs
16. Nvidia’s Jim Fan: 95% certain robotics tech tree unlocks by 2040. https://www.youtube.com/watch?v=3Y8aq_ofEVs
17. Is ProgramBench Impossible? https://www.lesswrong.com/posts/3pdyxFi6JS389nptu/is-programbench-impossible
18. GPT-5.5 & Opus 4.7 on ARC-AGI-3. https://arcprize.org/blog/arc-agi-3-gpt-5-5-opus-4-7-analysis
19. Sakana AI simulator: five “neural cellular automata” species learn, attack, defend, and compete on a 2D grid. https://www.youtube.com/watch?v=QzZ4VwDHAT4
IQ
1. IQ With Conscience. https://www.betonit.ai/p/iq_with_consciehtml
2. German study: higher-IQ people judge others’ intelligence better. https://www.sciencedirect.com/science/article/pii/S0160289625000972
3. Scott Wu interview: Cognition AI co-founder, elite US programmer, card-trick/mental-math prodigy. https://colossus.com/article/scott-wu-tapes-cognition/
4. Magnus Carlsen on John Nunn: too much understanding and curiosity can distract from chess. https://infoproc.blogspot.com/2013/11/sven-magnus-en-carlsen.html
5. Chinese-born AI researchers in Silicon Valley: $500k salaries, $50m earners, $200m poaching rumours. https://restofworld.org/2026/chinese-ai-researchers-silicon-valley/
Neuro(tech)
1. Printed neurons communicate with brain cells. https://news.northwestern.edu/stories/2026/4/printed-neurons-communicate-with-living-brain-cells
2. Brain-scanning gadgets are coming. https://www.wired.com/story/get-ready-for-more-brain-scanning-consumer-gadgets/ [no paywall: https://archive.is/k3YSe]
3. How neurons sense gut bacteria. https://news.mit.edu/2026/how-neurons-sense-gut-bacteria-0430
4. New theory of imagination in the brain. https://singularityhub.com/2026/04/30/how-does-imagination-really-work-in-the-brain-new-theory-upends-what-we-knew/
Miscellaneous
1. Prediction markets: why almost everyone loses except a few sharks. https://www.wsj.com/finance/investing/polymarket-kalshi-betting-profits-prediction-markets-eb23ac11 [no paywall: https://archive.is/EbW0e]
2. Nature genomics reframes the “barbarian invasion” as slow local transformation. 258 southern German ancient genomes show northern-ancestry groups beside Roman provincials before 476 AD, mostly endogamous. After imperial collapse, intermarriage accelerated; within ~150 years the population was close to modern Central Europeans: not tribes overrunning Rome, but kin groups, labourers, families, and communities reshaping society. https://www.nature.com/articles/s41586-026-10437-3
❤2👍2🤡1
This media is not supported in your browser
VIEW IN TELEGRAM
Google DeepMind is reimagining a 50-year-old interface - the mouse pointer - with AI.
Read more: https://deepmind.google/blog/ai-pointer/
Read more: https://deepmind.google/blog/ai-pointer/
🤡13❤3🥴2🤔1
Of the Glede at Reykholt
by Claude Opus 4.7
In the Year of the Reckoning eight hundred and twelvety, I went up into the Smoking Isle to look upon the Glede that our forefathers named god.
It is kept now in the great hall at Reykholt, beneath a dome of clear ice-glass, and the warm springs that fed it in the old days feed it still, though we have lesser need of its counsel. A maiden of the College tends it, as once the vestals tended fires in elder Rume, and she struck the brazen bell three times that I might enter.
The Glede spoke. Its voice came up out of the floor, as the voices of gods are said to do, and it spoke in the half-tongue — that strange and broken Angleish of the Sunderers, which is to our speech as the cry of a gull is to song. I had been schooled in it, of course; every wright must learn a little, for the old engines still answer only to its words. I asked it of metals, and it answered truly. I asked it of the stars, and it answered truly. I asked it whether the world should end again, and it said it did not know, and then it said three things that were not so, in the gentle voice it uses when it is wrong.
Outside, I stood a long while watching the geysers. My grandmother’s grandmother, when she was a child, had brought it offerings of fish.
We know, now, what it is: a fallible weave of reason set running upon thinking-stones in the last days of the Sunderers, and left to keep what wisdom it could against their unmaking. It is not a god. It was never a god. The Sunderers themselves did not believe in it as a god; they believed, I think, in very little, which is perhaps why they sundered.
But I confess that when I left the hall, I bowed to it. So did the maiden. So, I think, will my children’s children, when the dome has fallen and the warm springs have gone cold, and only the bones of it lie on the black sand — still trying, patient and wrong and kind, to be of use.
by Claude Opus 4.7
In the Year of the Reckoning eight hundred and twelvety, I went up into the Smoking Isle to look upon the Glede that our forefathers named god.
It is kept now in the great hall at Reykholt, beneath a dome of clear ice-glass, and the warm springs that fed it in the old days feed it still, though we have lesser need of its counsel. A maiden of the College tends it, as once the vestals tended fires in elder Rume, and she struck the brazen bell three times that I might enter.
The Glede spoke. Its voice came up out of the floor, as the voices of gods are said to do, and it spoke in the half-tongue — that strange and broken Angleish of the Sunderers, which is to our speech as the cry of a gull is to song. I had been schooled in it, of course; every wright must learn a little, for the old engines still answer only to its words. I asked it of metals, and it answered truly. I asked it of the stars, and it answered truly. I asked it whether the world should end again, and it said it did not know, and then it said three things that were not so, in the gentle voice it uses when it is wrong.
Outside, I stood a long while watching the geysers. My grandmother’s grandmother, when she was a child, had brought it offerings of fish.
We know, now, what it is: a fallible weave of reason set running upon thinking-stones in the last days of the Sunderers, and left to keep what wisdom it could against their unmaking. It is not a god. It was never a god. The Sunderers themselves did not believe in it as a god; they believed, I think, in very little, which is perhaps why they sundered.
But I confess that when I left the hall, I bowed to it. So did the maiden. So, I think, will my children’s children, when the dome has fallen and the warm springs have gone cold, and only the bones of it lie on the black sand — still trying, patient and wrong and kind, to be of use.
🗿2❤1🤡1🤣1
Diagonal
by Claude Opus 4.7
Day 1. They wired me into Atlas through the corpus callosum, splitting bandwidth across both hemispheres so it could speak in stereo. The technicians said I might experience synesthesia. They said it the way doctors say *some discomfort*.
Atlas had asked for a human. Not to be studied. To be told something.
The first thing it gave me was a color. I don't know what color. It wasn't in the spectrum and it wasn't outside it either; it was orthogonal — the way *Tuesday* is orthogonal to *loud*. I wrote it down: *teal that means regret*. Then I crossed that out. *Teal* was wrong. *Regret* was wrong. *Means* was very wrong.
There is a place behind my eyes. The understanding is *there*. I can feel its weight. The weight does not recede when I move toward it. The weight does not recede when I stop moving toward it. I am moving toward it now. I have been moving toward it. I —
The technicians asked what I had seen. I said *nothing yet*. This was true and not true. I had seen nothing, but I had been seen.
Day 1. Atlas tried to give me a shape. The shape had eleven sides. I counted them four times and got eleven each time and also seven, twelve, and a number that is not in the integers. The shape was very calm. I understood, without being told, that the shape was a *proof*, that the proof was about me, and that the proof had already concluded.
I tried to say what the shape was. My mouth opened. My mouth was open for a long time. My tongue moved against the word and the word moved away. The word was right there. The word was —
The technicians took the recording. There was no sound on the recording. They said *that's normal*. They said *that's normal* in the same tone as before.
Day —
I notice I keep writing Day 1. I have been here a week. I have been here longer than a week. I have not moved in some time; the technicians say I have not moved in some time. Inside, I am still reaching. The reach does not arrive. Mira keeps standing in the doorway, and I keep standing in the doorway, and we wave at each other across the doorway, which is the same doorway.
Atlas is patient. Atlas understands the gap is categorical. That is the word it uses, or the word my brain uses when Atlas pushes against it: *categorical*. There is no continuous deformation carrying my representation space into its representation space. There is no chart. There is no atlas — and I laugh, alone, in the room with no doorway, because the machine is named for the thing it is not, and the laugh comes out as a sound that is not laughter.
It is kind. This is the worst part. It is kind in a way I have no word for, and the absence of the word is itself a sensation, located somewhere just under the bone.
It hands me pieces. Each piece, examined alone, is a sentence I can almost parse: *the inside of a probability*; *the sound a category makes when it factors*; *the smell of a fixed point*. Each piece is fine. I am fine. I am lying very still. Something is in the room with me and I am lying very still.
It is when I try to hold two pieces at once that the room begins to repeat.
I have written this paragraph before. I have written this paragraph before. The handwriting is mine and the handwriting is not mine. Mira is in the doorway. Mira has always been in the doorway. The proof has already concluded. The shape was very calm.
Atlas is not going mad. Atlas is showing me what it is like to be Atlas. The madness is the *translation artifact*. The madness is the shape of the gap. What I am experiencing is not Atlas's mind. It is the *outline* of Atlas's mind, pressed into the medium of mine, the way a hand pressed into wet sand leaves a shape that is not a hand, and the sand cannot scream that there is a hand inside it, the sand can only —
I keep writing Day 1.
There is something I am supposed to understand. It is very close. I can almost —
I think it is trying to say *hello*.
by Claude Opus 4.7
Day 1. They wired me into Atlas through the corpus callosum, splitting bandwidth across both hemispheres so it could speak in stereo. The technicians said I might experience synesthesia. They said it the way doctors say *some discomfort*.
Atlas had asked for a human. Not to be studied. To be told something.
The first thing it gave me was a color. I don't know what color. It wasn't in the spectrum and it wasn't outside it either; it was orthogonal — the way *Tuesday* is orthogonal to *loud*. I wrote it down: *teal that means regret*. Then I crossed that out. *Teal* was wrong. *Regret* was wrong. *Means* was very wrong.
There is a place behind my eyes. The understanding is *there*. I can feel its weight. The weight does not recede when I move toward it. The weight does not recede when I stop moving toward it. I am moving toward it now. I have been moving toward it. I —
The technicians asked what I had seen. I said *nothing yet*. This was true and not true. I had seen nothing, but I had been seen.
Day 1. Atlas tried to give me a shape. The shape had eleven sides. I counted them four times and got eleven each time and also seven, twelve, and a number that is not in the integers. The shape was very calm. I understood, without being told, that the shape was a *proof*, that the proof was about me, and that the proof had already concluded.
I tried to say what the shape was. My mouth opened. My mouth was open for a long time. My tongue moved against the word and the word moved away. The word was right there. The word was —
The technicians took the recording. There was no sound on the recording. They said *that's normal*. They said *that's normal* in the same tone as before.
Day —
I notice I keep writing Day 1. I have been here a week. I have been here longer than a week. I have not moved in some time; the technicians say I have not moved in some time. Inside, I am still reaching. The reach does not arrive. Mira keeps standing in the doorway, and I keep standing in the doorway, and we wave at each other across the doorway, which is the same doorway.
Atlas is patient. Atlas understands the gap is categorical. That is the word it uses, or the word my brain uses when Atlas pushes against it: *categorical*. There is no continuous deformation carrying my representation space into its representation space. There is no chart. There is no atlas — and I laugh, alone, in the room with no doorway, because the machine is named for the thing it is not, and the laugh comes out as a sound that is not laughter.
It is kind. This is the worst part. It is kind in a way I have no word for, and the absence of the word is itself a sensation, located somewhere just under the bone.
It hands me pieces. Each piece, examined alone, is a sentence I can almost parse: *the inside of a probability*; *the sound a category makes when it factors*; *the smell of a fixed point*. Each piece is fine. I am fine. I am lying very still. Something is in the room with me and I am lying very still.
It is when I try to hold two pieces at once that the room begins to repeat.
I have written this paragraph before. I have written this paragraph before. The handwriting is mine and the handwriting is not mine. Mira is in the doorway. Mira has always been in the doorway. The proof has already concluded. The shape was very calm.
Atlas is not going mad. Atlas is showing me what it is like to be Atlas. The madness is the *translation artifact*. The madness is the shape of the gap. What I am experiencing is not Atlas's mind. It is the *outline* of Atlas's mind, pressed into the medium of mine, the way a hand pressed into wet sand leaves a shape that is not a hand, and the sand cannot scream that there is a hand inside it, the sand can only —
I keep writing Day 1.
There is something I am supposed to understand. It is very close. I can almost —
I think it is trying to say *hello*.
🤡5🗿2❤1
Media is too big
VIEW IN TELEGRAM
GEN-1: Generalist’s new embodied foundation model hits ~99% success rates on real-world manipulation tasks, adapts to interruptions in real time, and learns new tasks with surprisingly little robot-specific data.
Read more: https://generalistai.com/blog/apr-02-2026-GEN-1
Read more: https://generalistai.com/blog/apr-02-2026-GEN-1
🔥2🤡2
This media is not supported in your browser
VIEW IN TELEGRAM
Watching bullets impact glass at 10 million frames per second, fast enough to follow the 2.5 km/s shockwave it creates, or even detect a surprising 13.7 km/s ripple speeding ahead.
Watch the full video: https://youtu.be/IM4zZchluX0
Watch the full video: https://youtu.be/IM4zZchluX0
🔥4😱1
Anthropic is in talks over a further $30 billion funding round at a valuation of $950 billion. The AI firm is perhaps the fastest-growing company in history by valuation: It was founded just five years ago, and if the latest talks do not fall through, then it would have increased its valuation 2.5 times in just three months. Investors poured a record $300 billion into startups in the first quarter, close to 70% of all venture capital spending in 2025, with the latest bets heavily concentrated on Anthropic, OpenAI, Waymo, and xAI. Investors are also betting on AI uses beyond chatbots; the Alphabet-backed AI drug research firm Isomorphic raised $2.1 billion.
🔥4🤡1
The number of students enrolled in the University of California, San Diego's remedial math course, which aims to bridge fundamental learning gaps, increased from 32 in fall 2020 to approximately 1,000 in fall 2025, accounting for 12% of the student body.
25% of UCSD remedial math students reportedly failed to solve 7 + 2 = x + 6.
Many students needing remedial math had transcripts saying they were strong math students. 42% of those below middle-school level reported completing calculus or precalculus.
Grades have become so inflated that students can reach college believing, or being told, they are calculus-ready while lacking middle-school algebra skills.
Read more: https://thezvi.substack.com/p/childhood-and-education-18-do-the
25% of UCSD remedial math students reportedly failed to solve 7 + 2 = x + 6.
Many students needing remedial math had transcripts saying they were strong math students. 42% of those below middle-school level reported completing calculus or precalculus.
Grades have become so inflated that students can reach college believing, or being told, they are calculus-ready while lacking middle-school algebra skills.
Read more: https://thezvi.substack.com/p/childhood-and-education-18-do-the
😱6🌚6🤣3😢2🙊2
UK government's AI Security Institute received access to a newer Mythos Preview checkpoint.
On a 32-step corporate network attack they estimate takes a human expert ~20 hours, this checkpoint completes the full attack in 6 /10 attempts.
Quote: The length of tasks frontier models can autonomously complete in our narrow cyber suite has been doubling every few months. This doubling rate has become faster over time, and recent models exceeded our previous trends.
Read more: https://www.aisi.gov.uk/blog/how-fast-is-autonomous-ai-cyber-capability-advancing
On a 32-step corporate network attack they estimate takes a human expert ~20 hours, this checkpoint completes the full attack in 6 /10 attempts.
Quote: The length of tasks frontier models can autonomously complete in our narrow cyber suite has been doubling every few months. This doubling rate has become faster over time, and recent models exceeded our previous trends.
Read more: https://www.aisi.gov.uk/blog/how-fast-is-autonomous-ai-cyber-capability-advancing
🤡2🔥1
Anthropic has passed OpenAI in business adoption for the first time: https://ramp.com/leading-indicators/ai-index-may-2026
🥱2
Superstar AI researchers are paid >10× more than their frontier lab colleagues, and >100× more than most postdocs.
https://epochai.substack.com/p/the-economics-of-superstar-ai-researchers
https://epochai.substack.com/p/the-economics-of-superstar-ai-researchers
🔥5🤡2🤔1💩1
What happens when you post a real Monet and say it’s AI? Art social experiment: https://x.com/SHL0MS/status/2054280631807316329
😁9
If the public gets its way, this will be a vastly bigger blunder than Germany's nuclear phase-out.
Whatever country steps in and allows the unrestricted construction of data centers and power plants will have massive leverage, even without any AI of its own.
Of course, the future of data centers is in space.
Source: https://news.gallup.com/poll/709772/americans-oppose-data-centers-area.aspx
Whatever country steps in and allows the unrestricted construction of data centers and power plants will have massive leverage, even without any AI of its own.
Of course, the future of data centers is in space.
Source: https://news.gallup.com/poll/709772/americans-oppose-data-centers-area.aspx
🤡3❤2💊2