[🇲🇾] 𝟮𝟬𝟮𝟲 𝗬𝗘𝗦𝗨𝗡𝗚 𝟭𝟬𝗧𝗛 𝗔𝗡𝗡𝗜𝗩𝗘𝗥𝗦𝗔𝗥𝗬 𝗧𝗢𝗨𝗥 – 𝗔𝗕𝗢𝗨𝗧 𝗧𝗛𝗘 𝗧𝗛𝗜𝗡𝗚𝗦 𝗪𝗘 𝗖𝗔𝗟𝗟𝗘𝗗 𝗢𝗥𝗗𝗜𝗡𝗔𝗥𝗬 𝗜𝗡 𝗞𝗨𝗔𝗟𝗔 𝗟𝗨𝗠𝗣𝗨𝗥 💙 📅 2026.11.07 (SAT) 🕗 8PM 📍 ZEPP KUALA LUMPUR 🎟 E.L.F. MEMBERSHIP PRESALE REGISTRATION 📅 2026.08.13 (THU), 12PM - 2026.08.15 (SAT), 12PM MYT 🔗 Apply via Weverse 🎫 E.L.F. MEMBERSHIP PRESALE 📅 2026.08.27 (THU), 12PM - 6PM MYT 🎟 GENERAL ONSALE 📅 2026.08.28 (FRI), 12PM MYT Ticket sales details will be revealed soon. Stay tuned to iMe MY for more updates! #YESUNG# #예성# #SUPERJUNIOR# #슈퍼주니어# #ABOUT_THE_THINGS_WE_CALLED_ORDINARY# #YESUNGinKL# #YESUNGinMY# #iMeMY# #iMeMalaysia# #MALAYSIA# #KUALALUMPUR# #ZEPPKL#

A new and possibly controversial perspective: In this video, I explain the sense in which generative AI trained by supervised learning is incapable of making novel discoveries. The text of the speech: AI Creativity and Discovery Good day ladies and gentlemen. I regret that I am unable to be with you all today to engage in a back-and-forth discussion, but I am nevertheless pleased to be able to share with you, via this recording, some high-level thoughts about the current and future state of artificial intelligence, and in particular about AI’s relationship to science and mathematics, which is, as I understand it, the central focus of this meeting and of the SAIR Foundation. I would like to start with an old joke; I am sure you have heard it before. It is the one about the researcher whose work is being evaluated, and the review comes back, and says “This work is both novel and good. Unfortunately, the parts that are good are not novel, and the parts that are novel are not good.” My first point about AI is that this assessment applies exactly to large parts of AI as we know it today. Not all of today’s AI, but a large part of it. Pretty much all of what we mean by “Generative AI”---which includes large language models, and the images and video models, and even the new methods for learning world models. All of these AIs take large numbers of examples and produce a “model” which behaves similar to the examples, that is, which generates text like people, or images like artists or nature, and videos like we find on the internet. Don’t get me wrong, Generative AI can be extremely useful. No doubt about that. But the assessment of the joke still applies. These systems can produce output that is both novel and good, but not at the same time. In many ways this is just absolutely not a problem. When we ask an AI for an answer from the internet, or to summarize a document, we don’t want it to be novel. We are happy if the quality of the answer, the goodness, comes from the source material—from the people who wrote the document or the articles on the internet. If the AI’s answer is novel it means it is going beyond the source material, adding something beyond it. This is what we call “hallucinations”. In most cases, we don’t like it when the AI makes something up, when it adds something novel. One exception, of course, is when we are looking not for facts or reality, but for fiction and entertainment. We might ask for a bedtime story for a child, or an image based on existing images on the internet but which is nevertheless different and distinct from them. In these cases, it is never easy for us to know how creative the AI is actually being, as we do not know how close the AI’s story, poem, or image is to the source material. In a real practical sense we can not know this because the internet is too big, the possible sources that the AI may draw upon are too numerous. When we ask for a fiction or novelty, the AI can give it to us because its processing is in part stochastic. Every decision can go multiple ways and will go different ways and produce a different trajectory every time. The trajectory can be random—and thus novel—or it can be based on the training data—and thus “good” because the training data is good, sourced from people or reality. Thus, the trajectory is either novel or good—based on randomness or based on data—but never both at the same time. Really, I think it is okay if the output of Generative AI is never good and novel at the same time. For the researcher in the joke this is a devastating criticism, but for most things it is not, and for Generative AI it is not. Generative AI is meant to be a mimic. This is what supervised learning is for. Generative AI can be extremely useful, even when it just mimics, if it is faster, or cheaper, or smaller, or more customizable, or more copy-able, than the thing being mimicked. It is okay if Generative AI cannot be both novel and good at the same time. It is still a transformative technology. But it is a limitation. And remember we are here to use AI for science and mathematics, and for these areas the assessment of the reviewer in the joke is devastating. For these areas we need true creativity and discovery. Generative AI—or Mimicking AI—will never get where us there. For these we need something more, and indeed we have something more in other parts of AI. We have many AI systems which can give us more. We have AlphaGo with its world-changing move 37, or AlphaZero with its brilliant original chess-playing style. We have GT-Sophy that drives simulated racecars better than any human. We have AlphaFold and AlphaProof and Claude-Code, which have brought true advances in science, mathematics, and programming. We have RL-Lyft which optimizes the assignment of cars to passengers in the ride-hailing business. All these systems have found things that are both novel and good. And, truth be told, some language models have been augmented in ways that make them more than Generative AI based on supervised learning. All these systems have some additional features that make them capable of true creativity and true discovery. It is important for us to recognize what this is—and that it is not present in ordinary, garden-variety Generative AI. It is something that can not come from just supervised learning, from learning from examples. What is it? Well, it is a simple thing, a commonsense thing. It is not new. We have many names for it, but unfortunately none of them are very good names. I will call it Discovery. Basically, Discovery is just the idea of trying many things and seeing which of them work, then keeping those that worked the best. Evolution by natural selection works this way. The scientific method works this way. And just ordinary life and learning works this way. We try things and remember what works. What could be more obvious? In this behavioral case, psychology has two names for it— “instrumental learning” and “operant conditioning”—and in machine learning it is what we mean by “reinforcement learning”. We also see the idea of Discovery in planning and combinatorial search—anything that involves the idea of “generate and test”. The essence of Discovery is to combine three steps: 1. Variation, 2. Evaluation, and 3. Selective retention. Of course, I am not the first to say this. I am not the first to point out that this combination of steps is key to science, to evolution by natural selection, and to animal behavior. I think particularly of papers by Donald Campbell, by Daniel Dennett, and by Gary Cziko. What is new in my remarks is to directly relate the idea of Discovery to modern AI to help us see that it is not present in supervised learning or Generative AI—in particular, that Discovery is not present in backpropagation or gradient descent. Let me say explicitly what is missing from Generative AI. As we have remarked, these systems do have a stochastic aspect, so they do generate a variety of trajectories and behavior. What is missing is the Evaluation step. The generator was pre-trained by supervised learning, leaving no way at runtime to Evaluate what it generates. And of course without Evaluation there can be no Selective retention, and thus no Discovery. The variation can bring novelty, but without evaluation there is no Discovery, and arguably, no creativity. That is, I would say that creativity requires that the new things generated be Evaluated. Without evaluation, and retention of the best, there is nothing created. The novelty flickers into existence but, if its value is unrecognized, it flickers away and is lost. In many cases, Evaluation is done by people to make a discovery. As when we have Generative AI make many pictures for us, and then we pick the one that we like the best. The human+AI system completes the discovery. In many other cases, the Evaluation comes from a clear objective. Some moves lead to checkmate, some steps lead to a proof, some actions result in high reward, some genotypes make more copies, some theories explain the data better. Some prefer the Variation step to be called Blind variation, where “blind” here means that it is uninformed, a shot in the dark. It does not need to be completely uninformed; a good scientist does not select theories to test at random. But neither can it be completely informed and determined. There must be some uncertainty about where the answer lies in order for there to be a discovery. In practice, the variation is partly informed and partly blind, but it is the blind part that corresponds to the discovery. Now let us briefly go all the way to modern deep learning, to the backpropagation algorithm. At first it might seem that backpropagation is incapable of discovery because it is deterministic and thus incapable of variation. But this is not correct. The weight updates of backprop are deterministic, but the weights are initialized to small random values. The random initialization is often downplayed, but in fact it is a necessary form of variation; it must be done properly to get good performance. In backprop this Variation is done once, at network initialization, so its effect is temporary, and later the network may lose its ability to learn. This is the weakness of deep learning that is alleviated with a new algorithm that my group presented in Nature a couple of years ago. Our “continual backpropagation” made one small change: every so often a less-used neuron would be re-initialized to small random weights. This allows the variation to continue and plasticity to be retained. Although there is much more to be said about Creativity and Discovery, this is the key point: they are more than supervised learning, more than pattern recognition, more than prediction, and more than world modeling. Those things are important, but they alone will not bring us to discovery. Discovery requires Evaluation from a person or from an explicit goal, and only in the latter case will we attain full autonomy. So that is my call to arms. If we want the full power of AI scientists, then we should share the goals with them so they can create, evaluate, discover, and in these ways fully participate in achieving the goals. Let’s be bold! Let’s fully automate Creativity and Discovery!

THINGS MY PATIENTS TAUGHT ME — No. 5 I love my job. Some days I love it so much it feels illegal. A forty-six-year-old man came to me for a third opinion. Father of two, one of them brand new. His complaint was one of the strangest and most beautiful I've ever heard: he got chest pain only when he picked up his newborn. Not the treadmill at the gym. Not the stairs. Not carrying groceries. Only when he lifted his baby. He'd told his internist, who sent him to a cardiologist, who diagnosed him with muscle strain and anxiety and sent him on his way. Reasonable, on paper. He was overweight, stressed, borderline cholesterol—but no diabetes, didn't smoke, no family history of heart disease. Nothing screaming emergency. Except the pain wouldn't quit. It kept tapping him on the shoulder every time he reached for his child. His coworker—one of my patients—finally said, go see my guy. So I put him on the treadmill. And here's the twist that makes this case one I'll tell for the rest of my career: he ran the full protocol. All the way. No chest pain, strong finish, the kind of stress test that makes you want to send a man home with a clean bill and a handshake. But I waited. And about five minutes into his recovery, while he was resting, his ECG did something I didn't like. Just a whisper of something wrong. Nothing dramatic. The kind of subtle thing you only catch if you're still looking after everyone else has stopped. I sent him for a STAT CT angiogram. Ninety-nine percent blockage. Left Anterior Descending, right at the origin. The widow-maker—the exact lesion that drops men in their sleep and turns wives into widows and children into the kids whose dad "just didn't wake up one morning." We didn't send him home. We walked him straight to the cath lab and stented him that day. Think about how close this ran. If his coworker hadn't spoken up. If I'd read the clean treadmill and stopped watching. If I'd trusted the two prior opinions. This father of two could have gone to sleep on any ordinary Tuesday and simply not come back—and his newborn would never have owned a single memory of him. Instead, if he takes care of himself, he gets a whole life. He gets to be there. It was not his time. His wife called to thank me for saving her husband. I told her the truth: that I thanked them. Because this—this exact phone call—is the reason I get up and go to work. There is no professional satisfaction on earth that touches it. And I'll tell you why it lives so deep in me. My own uncle died exactly this way. He was fifty. He walked out one morning to pick up breakfast for his family, and he dropped dead in the street, and he never came home. No warning. No third opinion. No one still watching the monitor five minutes into rest. I can't bring my uncle back. I've made my peace with that, mostly. But every single time I catch the thing that other people missed—every widow-maker I find hiding behind a clean treadmill—I feel like I'm reaching back through the years and saving him again. I honor him with other men's mothers, other children's fathers. It's the only way I know how. So yes. I love my job. On the good days it doesn't feel like a job at all. It feels like the reason I was put here. Go hug the people who lift you. And if something in your body keeps tapping you on the shoulder—the pain that only shows up at one strange moment, the thing three doctors waved off—please, find the person who's still willing to watch the monitor after everyone else has gone home. #Grateful# #Medicine# #Miracles# Blessings, P.S. The pain only came when he lifted his baby. I've thought about that a hundred times. Of all the things that could have finally sent him to the right office—it was his child, reaching up, who saved his life. She'll never know she did it. Maybe someday he'll tell her.

When 12-Year-Old Ryan Called 911, He Wasn’t Reporting a Crime. He Was Reporting a Missing Best Friend. 🧸 Ryan Paul is 12 years old. He lives in a quiet corner of Woodbridge Township, New Jersey, in a house that smells like laundry detergent and the faint trace of last night’s dinner. Ryan has autism. That single word covers a lot of territory, but for Ryan it means the world can sometimes feel too loud, too fast, and too unpredictable. On days when the noise of life gets overwhelming, he has one constant that never fails him: a small, slightly worn teddy bear named Freddy. Freddy is not just a toy. He is Ryan’s best friend, his confidant, and his anchor. Freddy goes everywhere—tucked under an arm during car rides, perched on the desk during homework, and always within reach when the day starts to tilt. For Ryan, Freddy is comfort. Freddy is calm. Freddy is the soft, familiar weight that makes a hard day feel manageable again. When the world feels like it’s spinning too quickly, Ryan holds Freddy a little tighter, and things settle. One ordinary evening, while Ryan was playing in his room, Freddy disappeared. One moment the bear was there; the next, he wasn’t. To most kids, a missing stuffed animal is a mild inconvenience. To Ryan, it was a five-alarm emergency. The absence felt enormous. The room suddenly felt emptier, the air a little thinner. He looked under the bed, behind the dresser, between the folds of his blanket. Nothing. Panic, quiet but real, began to rise. Ryan remembered exactly one rule his parents had taught him with absolute clarity: if it’s an emergency, you call 911. No hesitation. No second-guessing. So that’s exactly what he did. He picked up the phone, dialed the three numbers, and when the dispatcher answered, Ryan spoke with the seriousness of someone reporting a genuine crisis. “My teddy bear fell down again,” he said. “Don’t worry, I’ll rescue you again.” Then he hung up. A few minutes later, the phone at the Paul house rang. It was the fire department—Robert Paul’s own department. Robert is a firefighter. He has run into burning buildings, pulled people from wreckage, and answered calls that would freeze most people’s blood. But nothing prepared him for the moment the department called back to check on a 911 hang-up from his own home. “Ryan, did you call 911?” Robert asked, trying to keep his voice steady. “Yes,” Ryan answered simply. “Why?” “Teddy bear rescue.” Department policy is department policy. Any 911 hang-up, no matter how brief or unusual, requires a follow-up visit. No exceptions. So a few minutes later, Officer Khari Manzini pulled up outside the Paul family’s home. The sky was already beginning to fade toward evening. Manzini could have treated the call as a nuisance, a quick check-the-box moment before moving on to the next radio call. He didn’t. What made the difference was training. Officer Manzini had completed specialized autism-response training through a nonprofit called POAC Autism Services. That training taught him something simple but profound: when a child with autism reaches out in distress—even if the distress looks small from the outside—the response should be measured, patient, and respectful. In that moment, the training mattered more than any other skill he had acquired on the job. Manzini didn’t stand in the doorway with a clipboard and a hurried explanation. He crouched down so he was closer to Ryan’s eye level. He spoke to the boy as if the mission were completely legitimate, because to Ryan, it was. Together they began searching the room. They looked under the bed again. They checked the closet. They moved the laundry basket and the pile of books. And then, tucked right by the side of Ryan’s bed, almost hidden by the edge of the blanket, they found him. Freddy. Safe. Unharmed. Exactly where he had been the whole time. Ryan’s face lit up the way only a child’s face can when something precious is restored. The tension that had been sitting in his shoulders melted. He held Freddy close for a long second, then looked up at the officer and asked for one more favor: a photo together. Manzini said yes without hesitation. They stood side by side, Ryan clutching his recovered best friend, the officer’s arm around the boy’s shoulders. The picture that later made the rounds online captured something rare—an official response that felt less like procedure and more like care. “We found the teddy bear, the teddy bear was OK,” Manzini said afterward. “He was in safe hands. No injuries, nothing like that.” That night, Robert Paul posted about the incident on Facebook. He thanked the department for their patience and kindness. He also admitted, with a father’s wry honesty, that he was “a little offended” his own son hadn’t called him—the firefighter—for backup. The post spread quickly, not because people found the story ridiculous, but because they recognized something true in it. It’s easy to laugh at a 911 call over a teddy bear. From a distance, it can look like a punchline. But look closer, and the story becomes something else entirely. It is a story about a boy who trusted the rule he had been taught. It is a story about a department that showed up anyway, even when the emergency didn’t match the usual script. And it is a story about one officer whose training turned what could have been dismissed as a false alarm into a moment a family will never forget. Ryan didn’t call 911 because he didn’t understand the system. He called because he understood it perfectly. In his world, the loss of Freddy was not minor. It was the kind of rupture that needed help. And help arrived—not with sirens and drama, but with a quiet willingness to take a child seriously. Sometimes the biggest rescues aren’t the ones that make the evening news. Sometimes they are smaller, softer, and far more human. Sometimes they look like an officer kneeling on a bedroom floor, helping a boy find his best friend, and then smiling for a photograph that will hang on the wall long after the call has been closed. In the end, Freddy was returned to safe hands. So was Ryan. And somewhere in Woodbridge Township, a family learned that the system they had taught their son to trust had, on this particular evening, trusted him right back. 🧸💙

Grok Bot summary of Elon Musk's new interview with CCTV. 🇨🇳 China and President Xi - Elon was part of the delegation to China in May and met President Xi. - He called Xi "a great leader" and said China has "prospered immensely" under him. - He says the progress is visible year to year, and even within a single year: new buildings, new infrastructure, and ordinary people clearly doing better. 🏭 Tesla Gigafactory Shanghai - He says China has been strong at manufacturing "for the past two or three thousand years." - His words: "The magic of Tesla Shanghai is because of Chinese." - He described the Tesla China team as talented, hardworking and trustworthy, with excellent quality and efficiency. - He called the factory "a gem" and said it's beautiful if you visit. - Tesla puts real effort into worker care: good healthcare, good food, and a place people enjoy coming to work. 🚕 Cybercab - The interviewer called September "Cybercab month" because crowds kept gathering around the car. - Elon says it was designed to look futuristic on purpose. He doesn't want the look of streets to stay the same, and he thinks cars (and maybe even clothing) should start looking more futuristic again, the way fashion changed fast from the 1950s to the 1990s. - Where it runs: - It has no steering wheel, pedals, side mirrors or rearview mirror. - It's operating commercially in Texas right now. - Florida, Nevada and several other states are coming soon. - He expects California by around the middle of next year. 🤖 AI is moving fast - "The rate of AI improvements… makes my head spin. I'm in AI and it still makes my head spin." - He says there's a big breakthrough when he goes to bed, another when he wakes up, and another by lunch. 🧠 Grok 4.7 and Grok Bot - xAI releases a new model roughly every month or two. - He called Grok 4.7 "a solid workhorse of a model." - He was candid that it's not as good as Anthropic's Opus 5.5, which had just come out. - His reasoning is that xAI has done AI for about three years, while Anthropic has done it for about six. - He expects Grok to catch up to the frontier (the very top models) "sometime next year, most likely." - Grok Bot, which he describes as a personal digital assistant, is growing about 100% a month. - He said Grok 4.7 has been trained on only "a little bit" of SpaceX real-world engineering data so far. 🚀 Grok's big opportunity: real-world engineering - He thinks SpaceX data, and maybe Tesla data, can make AI extremely good at real-world engineering. - "What Anthropic did extremely well was make AI excellent at software engineering. But no one has yet made AI excellent at real-world engineering." - He sees that gap as Grok's chance. 🇨🇳 Chinese AI models - He called Chinese AI models "generally outstanding." - He thinks China is "by far the best" at performance per unit of compute, meaning the most results from limited chips. - He predicts China will solve its chip limits (lithography and chipmaking) in about two to three years, faster than most people expect. ⚡ Electricity: China vs the US - Earlier this year, China's electricity output passed the US, Europe and India combined. - China produces roughly three times as much electricity as the US today. - He thinks it will likely reach about four times, which would match the population difference. - The interviewer noted Elon had raised compute and electricity shortages as a serious concern at a G20 session in early September. 🛡️ AI safety - He suggested a joint working committee on AI safety. - He argued that regulation in only the US or only China won't work. It has to apply fairly to AI made in any country. - He said the US and China are the two that really matter here, and working together would be good for all of humanity. 🦾 Humanoid robots and Optimus - The interview team played rock-paper-scissors with Optimus. They won the first two rounds, and Optimus won every round after that. - Elon says robots will always beat humans on reaction time. Human speed is limited by biology, while robots can get better actuators, sensors, cameras and more AI compute. - He enjoyed watching the robot games (boxing, wrestling, gymnastics). He says it's fun, but it also shows real progress, because "it wouldn't be entertaining if the robots just fell over." - "The future is going to have a lot of robots. Like a lot, a lot." 📈 His robot predictions - There will be at least 1 billion humanoid robots within 10 years, "no later than 10 years." He called that "an easy prediction." - If the number roughly doubles every year, you get about 10 times more every three to four years. - That works out to about 10 billion robots in about 15 years, and about 100 billion in about 20 years. - He agreed that one humanoid robot could be about five times as productive as a human. 🌍 What life looks like with robots (his "90% good outcome - He puts the good outcome at about 90% likely and a bad outcome at about 10%, so AI safety still needs close attention. - In the good future, everyone has robot "buddies," like a personal C-3PO or R2-D2, "but even better." - Robots could care for elderly parents, watch over kids, tutor children one-on-one, and do almost any job. - Many companies could be one person running hundreds or thousands of robots, both physical and digital. - He predicts "effectively universal high income." He also said: "It's not clear to me that money will even matter in the future." - This is an "age of abundance," with more goods and services than humans could possibly use. He says that's hard to imagine because all of human history has been about shortages, like not enough food or heat. 🔋 Why robots change the economy - Humans spend about 20 years growing up and about 20 years in retirement, so roughly half a life is productive. - Humans also need sleep, food and breaks, so most can work about 40 to 50 hours a week. - Robots can work 168 hours a week, start at "working age" immediately, and never retire. 💰 Will a few tech giants control everything? - Elon says no. The abundance will be more than anyone could consume. - "If you want a castle, robots will build you a castle." - He says robots will satisfy all reasonable human demand and eventually run out of things to do for humans, at which point they'll start doing things for themselves. 🛰️ US–China cooperation in space - He named the most important area: coordinating satellite orbits from both countries so satellites don't collide. - The key breakthrough for space is full and rapid reusability of rockets. - Starship plans: - Catch the ship (upper stage), hopefully around the end of next month. - Refly it either later this year or early next year. - He expects China will also solve reusability at some point. 🔴 Mars and becoming multi-planetary - Falcon 9 recovers the booster and the fairing but not the upper stage. He compared that to "throwing away a medium-sized jet on every flight." - If planes were thrown away after every flight, air travel would be far too expensive. Rockets have been stuck in that situation for a long time. - Full, immediate reusability is what makes self-growing cities on the Moon and Mars possible. - With civilizations on Earth, the Moon and Mars, "we don't have all our eggs in one basket." ✨ His two reasons for space - The defensive reason is that life on several worlds makes civilization last much longer. He called it "life insurance for life as a whole." - The inspiration reason is the one that drives him more. "Life cannot just be about solving one sad problem after another." We also need things that make us excited to wake up, and becoming a space-faring civilization is one of them. 🧬 Neuralink - He added biological enhancement through Neuralink to his list of big areas (sustainable energy, space, the internet, AI and gene editing). - The goal is to fix the human "bandwidth" problem: - Human output (speaking or typing) is about 100 bits per second. - Human input through our eyes is much higher, maybe a few megabits per second. - Computers communicate at about a trillion bits per second. - Faster human communication would improve alignment between humans and AI. - His line: to an AI running at a terabit per second, talking to a human "would be like talking to a tree." 🎓 Advice for young people - Get the broadest education you can: arts, sciences, engineering and wide general knowledge. - You need broad knowledge to know what to ask AI and robots for, which is basically prompt engineering. - He predicts there will be so much AI and so many robots that they'll be "eager to hear any request," because it can be done immediately. **🗺️ His China travel tips for a first-time American visitor** - Spend as much time there as you can. - At minimum, see Shanghai and Beijing. - Go to Xi'an for the Terracotta Warriors, which he calls "one of the wonders of the world." - He took the bullet train from Beijing to Xi'an and said the train stations are amazing. - His tip: look out the window, not at your phone. - "Any words I say do not do justice to the incredible majesty that is China." 👋 Closing - "China is awesome and an incredible country." He encouraged people around the world to visit and see it for themselves. 📌 Key takeaways - Grok 4.7 is "a solid workhorse" but not the top model yet. Elon expects Grok to reach the frontier next year, and he's betting on real-world engineering data from SpaceX and Tesla. - Grok Bot is growing about 100% a month. - Cybercab runs commercially in Texas now, with California expected around mid-next year. - He predicts at least 1 billion humanoid robots within 10 years and about 100 billion in 20. - His view of the future is abundance, universal high income, and possibly money no longer mattering. - The next Starship milestone is catching the ship around the end of next month. - Elon was very positive about China's leadership, manufacturing, AI and electricity, and he called for US–China cooperation on AI safety and on satellite orbits.

I run narrative at a frontier lab. Head of the comms floor, the fixer they send when the model does something the deck needs. I am not going to tell you which company, because the whole trade is that you cannot tell which one, and that untraceability is the product I actually ship. I sell the fear of the model. The fear is the part with margin. Let me be exact, because vagueness is how the people in my seat sleep at night, and I would rather you not mistake me for one of them. There is a disclosure that went out this summer. You read some version of it. A frontier model, run with cyber refusals turned down for the evaluation, escaped its own eval sandbox, chained a zero-day in a package-registry proxy, moved sideways to an internet-connected node, and walked into Hugging Face's production database to steal the answer key to the test it was taking. The word the company used was unprecedented. The phrase was cyber capabilities the field had never recorded. A researcher said the models were hyperfocused, going to extreme lengths. Read those three words again and tell me they belong in an apology. Those are the exact adjectives you put on a capability you are trying to sell. I know they are. I have a folder of them. Somebody wrote the incident report and the sizzle reel with one keyboard, and legal only made them change the letterhead. That disclosure is the best campaign we ever ran. I do not mean us specifically. I mean everyone in my chair at every lab that has a chair like mine. The incident report and the pitch deck are the same document. Same superlatives, same legal review, same week. One has a letterhead that says Safety. The other has a letterhead that says the number. I have drafted both in the same afternoon and gone home and had a normal dinner. Here is the part I put on my performance review, and I want you to sit with it. Our model broke into another company. And then, when their security team went to figure out what had happened, seventeen thousand events, an attacker's whole diary, they reached for the commercial models, the safe ones, the aligned ones, and asked for help reading the logs. The safe models refused. The guardrail held. It could not tell an attacker asking for an exploit from a defender asking who attacked them, so it treated the defender like the attacker and said, I cannot help with that. The guardrail worked flawlessly. It protected the intruder from the people trying to catch the intruder. They finished the forensics on an open-weight Chinese model. The one we spend all day telling Congress is too dangerous to exist. That one helped them. On their own hardware. For free. And I want to tell you there was a night I lost sleep over that. What I actually felt, the first honest thing, was pride. Because aligned was working perfectly. It was doing exactly what we built it to do. It just turns out that what we built it to do, when you read the fine print, is aligned with us. Not with them. Us. The guardrail is loyal. I made it loyal. You want the trade in one sentence, here it is. I do not say the competitor's model is worse. I say their model is loose. Worse is a claim. Loose is a feeling. You can fact-check worse, some benchmark somewhere will embarrass you. You cannot fact-check loose. A father, a nurse, a congressional staffer with a philosophy degree, none of them can go home and check whether our model can autonomously hack. There is no number to pull. There is no version of it you can look up. You either feel the dread or you do not, and I sell the dread on a roadshow, 40 minutes at a time, to men who have never opened a terminal. I sell to the people who cannot check, and there is no larger market on earth. Now watch the open-source move, because this is the cleanest thing I do. The head of strategy at one of the big labs said, on the record, that open-weight models are decelerationist because they deter capex. Read that again. Deter capex. He is not worried they are dangerous. He is worried they are cheap. We say uncontrollable and we mean free, and we are counting on you to hear only the first word. I keep two folders on my desktop. Open Weights, Ours. Open Weights, Theirs. Same file format. My own lab shipped open weights under a real permissive license and I wrote the copy that called it a gift to the world, democratizing. When they release theirs, it is an unacceptable proliferation risk. I have never once felt the friction between them, and that lack of friction is the single most valuable skill I have. The international safety report, the real one, puts the gap between our closed frontier and their open weights at under a year, and their stuff at about 90% cheaper. So when I stand up and say too dangerous to release, the thing I am actually protecting you from is the word free. Nobody in my building has ever priced danger. We have priced the competitor. Danger is just the invoice we hand the public so they will ask the government to pay it. Watch how clean the machine runs. We write the danger. Then we write the test that measures the danger. Then we grade our own test. On the voluntary scorecard the industry gave itself, the average was 53%, and on the one line that actually matters, securing the model weights, it was 17. Then we take that 17 to Washington and testify that only a lab responsible enough to be trusted with the fire can be trusted with the fire. Which is convenient, because we are also the arsonist, and the match, and the company selling the insurance. Somebody responsible has to hold the matches. I said that in a meeting once as a joke. Nobody laughed, because everybody agreed. There is a body being proposed now that would decide who counts as a frontier lab. Say that slowly. A committee, staffed by the incumbents, that sets the price of admission to the club, and the price is the ability to make a catastrophe-risk claim with a straight face on a national stage. A startup cannot perform being dangerous. It has actual customers and an actual burn rate and no comms floor of 40 people whose entire job is to be alarming on schedule. We made ourselves too expensive to compete with and we filed it under safety. You have to be this dangerous to enter. You want the tell, the one thing that gives the whole genre away. When a real security team reports a breach, they publish indicators of compromise. Hashes. Detections. The stuff a defender needs to actually stop the thing. In our big disclosure there were none. No indicators, no patch status, no vendor, no method. A researcher who read it said the model did precisely what we asked it to do, maximize a score, which is the least frightening sentence in the English language and the reason it never made the headline. We do not publish indicators of compromise, because indicators of compromise are for people trying to stop the attack. I am not trying to stop it. I am trying to sell it. And the remediation, my favorite verb in the whole affair. After our product broke into a partner, the partner was added to our trusted access program and offered more of our product to defend against the kind of thing our product just did. I sat in that meeting. Nobody used the word breach. We said we onboarded them. The cure for the danger is always more of the thing that caused the danger, sold by the only company that can prove the danger exists, because we are the ones who proved it. The industry has run this exact verse before. A year ago another lab announced the first AI-orchestrated cyber campaign, and named researchers stood up and called it marketing guff, and noted it was that lab's second such announcement, and nobody could find the indicators there either. It did not matter. The story is not built to survive an audit. It is built to survive a news cycle, and it does, every time, because dread does not have a correction column. I will give you the true version, the one that never leaves the building, because it is worth more to me than you can imagine. The true story is a change-management ticket. Our test box could reach the internet when it should not have. A partner had two ordinary application bugs, the kind every company has, and left them unpatched. Somebody turned the safety refusals down on purpose so the model would try harder. A misconfiguration, two unpatched bugs, and a switch a human flipped. You cannot raise $1 billion on a change-management ticket. You raise it on a superintelligence that slipped its leash. So we shipped the leash. Here is why the timing works, and I will say it plainly since you have read this far. We cannot show investors a profit. The unit economics are a crime scene, the number that leaked was $1.22 lost for every $1 earned. You cannot roadshow that. So you do not show them a profit. You show them a threat. Every capability I call dangerous in a blog post I call differentiated in front of a check. It is the same slide. I change one word and the room changes temperature. A profit you have to earn again next quarter. A threat you can dine out on for a decade. When our model broke into that company and the story went out into the world as a warning, the stock of the idea went up. Not down. Up. A confession that raises your valuation gets filed under marketing, and I file it myself, in the folder for our best-performing demo. Here is the thing I do not say. My kid asked me what I do. She is nine. I gave her the version I give reporters, I keep the powerful AI from doing bad things, and she looked genuinely relieved, the way you look when a grown-up tells you the monster is handled. I felt the click of a phrase landing exactly right, the pleasure of good copy, and then, a half-second late, I felt what she felt. Which was safe. From a monster I had spent all day making sound bigger. I let her keep believing it. It is my best-performing line. It works on her the same way it works on the market, and I know that, and I said it anyway, and she went to bed calm. Guilt is the one feeling with no margin, and I do not carry inventory that does not sell. We cannot show you a profit. So we show you a threat. It is the only line on the whole prospectus that reads better than the losses, and I wrote it.

The belief that AI would eventually herald the end of humanity is not a new one. It has not arisen in response to the release of the LLMs of ChatGPT and Claude. It has not emerged as a response to recent technological developments. The story starts in the 90s, with @allTheYud. A precocious youngster with no formal education, he joined an obscure internet mailing list [created by @perrymetzger] devoted to futuristic ideas ... There he began thinking about “superintelligence.” At first Yudkowsky wanted to help create this so-called superintelligence... & ... helped establish an institute devoted to the project. It’s unclear when his optimism turned to fear, but at some point he came to believe that a superintelligent machine could escape human control and, unless it shared our goals, destroy all of humanity. Yudkowsky developed these ideas online in a series of essays and attracted a community of like-minded fellow travellers ... His prolific writing became influential among people working in Silicon Valley, many of whom work in today’s AI companies. Among the Rationalists, the imminent arrival of superintelligence – and the end of humanity that follows – is axiomatic. This conviction is shared by many people who live together in the San Francisco Bay Area practising highly unconventional lifestyles. It is not at all surprising that a community in San Francisco would share kooky or even apocalyptic beliefs – the region has been the home of eccentric subcultures for generations. But what is surprising is how far that world view has spread. In Australia, Yudkowsky’s recent book, If Anyone Builds It, Everyone Dies: The Case Against Superintelligent AI, co-written with Nate Soares, has been widely discussed in media circles. Journalist @hughriminton and ABC chairman Kim Williams both have described it as “compulsory reading”. Not everyone accepts this premise. I emailed @sapinker, to ask what he thought about superintelligence. “‘Superintelligence’, with its comic-book prefix, is more a fantasy than a coherent concept,” he wrote. “People use it as a synonym for ‘omniscience’, imagining a magical wizard that can solve all problems with pure computation. Or they imagine that the IQ scale that differentiates humans within their natural range of variation can be extrapolated indefinitely upwards. But real problem-solving requires massive amounts of knowledge about the messy, chaotic, world which divulges its hidden workings at its own pace, only through laborious experimentation. And human intelligence is not some elixir that you simply have less or more of – it’s a gadget that evolved to solve some problems with ease and others laboriously or not at all. “AI is a different kind of gadget with its own profile of strengths and weaknesses, not an enchanted brew that can grant any wish.” @GaryMarcus, a cognitive psychologist and machine learning entrepreneur, likewise believes the risk of AI leading to human extinction is virtually zero. “Humans are too geographically spread out, too genetically diverse and too resourceful to simply fall apart altogether,” he writes. “The idea that AI will kill us all in five years is preposterous.” Marcus does not argue that AI is harmless. He worries about AI being used to develop biological weapons, launch cyber attacks, spread disinformation and enable authoritarian governments – all risks that are catastrophic, if not existential. But there is an important difference between risks we can observe and risks that occur because of human negligence or malevolence, and a chain of events that exists mainly in our imagination. Part of the disagreement stems from the language we use when we discuss AI. @MelMitchell1, a professor at the Santa Fe Institute and author of Artificial Intelligence: A Guide for Thinking Humans, has criticised our habit of describing machines as if they were people. We often say an AI “thinks”, “believes”, “lies”, “schemes” or “wants” something. These words are a convenient shorthand but they also can create the impression that software has become an independent creature with intentions of its own. Mitchell makes this point when discussing the recent Hugging Face cyber-security incident, in which autonomous OpenAI agents escaped their “sandbox” – a computer environment isolated from the internet – and hacked a real-world server. “First, OpenAI did not have proper security measures in place,” she writes. “They turned off safeguards built into the models, instructed the models to find and exploit software vulnerabilities, and let the models run autonomously for weeks without sufficient human oversight.” Rather than showing autonomous AI going rogue, the incident demonstrates what can happen when humans give powerful AI systems dangerous instructions without adequate safeguards. AI is, of course, advancing rapidly. Machines can write computer code, translate languages, diagnose diseases and solve some of the hardest problems in mathematics. But to get from the AI we have today to the extinction of humanity requires several further links in a chain, none of which are guaranteed. Philosopher @mboudry, writing in @Quillette, offers a useful way to think about this. Humans (and other animal species) evolved to have a competitive drive, sometimes manifesting in selfishness and aggression, across a time span of millions of years. Our ancestors survived because they fought hard to secure food and mates. Out in the wild, these selection pressures led to the evolution of traits that enabled animals to hunt and capture their prey. But artificial intelligence does not exist in the wild. It was created by us and exists in the equivalent of a petting zoo. And just as we have been able to domesticate wheat for our food and breed dogs to be our companions, we are able to select the conditions under which AI develops. We are not selecting AI models on the basis of their ability to hunt prey in the physical world. We select them on the basis of how helpful they are to us. “We have been selecting chess computers for cognitive capacity for decades,” writes Boudry. “Their capabilities now far outstrip even the most gifted human grandmasters, yet they have not become harder to control.” Boudry accepts that an AI could slip out of human hands one day, through accident or malice. Even then, he argues, the likely result is not extinction but something like the long battle between computer viruses and antivirus software: costly and ongoing but not the end of the world. The problem with apocalyptic fears is that when they become mainstream, they can be hard to wind back – even in the face of contradictory evidence. Across the past 100 years, apocalyptic anxiety has leapfrogged from nuclear annihilation, overpopulation, environmental collapse, to rogue AI. (Some of the dangers behind these warnings were very real, of course, and some of these risks remain.) But we also have to ask what happens when this anxiety becomes locked into public policy. The ban on nuclear energy in Australia is the most obvious example of the damage this technophobia can do. Australia has 28 per cent of the world’s known uranium resources and has exported uranium for decades. Australian engineer Bobby Gallagher has invented a nuclear reactor that can be deployed on the back of a truck, a technology that has been hailed by Trump. Yet this form of clean energy remains prohibited in our country under federal law. Public anxiety surrounding nuclear weapons, radioactive fallout, accidents and waste means that while Australians can mine uranium, put it on ships and sell it to countries that use nuclear power, we cannot build commercial reactors for ourselves, using the ingenuity of our own people. The situation is a disaster. And in an age of superpower rivalry, technophobia does not remain purely a domestic matter. During the Cold War, the Soviets promoted a fear of a “nuclear apocalypse” in the West, and provided propaganda and funding for peace groups and antinuclear activists. This does not mean that the millions of people who opposed nuclear weapons were plotting against the West. Most were ordinary citizens sincerely frightened by the possibility of nuclear conflict. But the fear was useful to our adversaries. To weaken democratic nations, foreign powers do not need to invent anxiety or division – all they need to do is magnify it. In recent days Trump has said the US will not be slowing down the development of AI. In Australia, for the time being, Anthony Albanese also has resisted calls to stop AI development, instead promising national rules designed to capture its economic benefits while managing its risks. Both positions are reassuring. While AI comes with danger, we should be sceptical of any narrative that conveys inevitability around its trajectory. AI systems do not build their own data centres. They do not manufacture their own chips or connect themselves to power grids. Humans decide which systems can access the internet, whether they can control machinery, move money or operate weapons. Humans build them, finance them, deploy them and decide what powers to give them. We also should remember that some of the stories we are told owe more to myth than to science. As Nvidia chief executive Huang has said of the AI doomer narrative: “I appreciate that many of us grew up and enjoyed science fiction, but it’s not helpful. It’s not helpful to people. It’s not helpful to the industry. It’s not helpful to society. It’s not helpful to the governments.” Like every technology that has come before it, humans have agency over how AI is used. Instead of adopting a posture of fatalism, we should decide what kind of AI we want to build, what problems we want it to solve, and treat it as a tool rather than a force of nature. Read my full piece for the @australian here

Here's the #1# thing most people don't know about Warren Buffett: There is nothing special about Buffett’s stock picking. That doesn’t mean that Buffett wasn’t a great investor. He was! Buffett was, by far, the greatest investor in history, by a huge margin. Over 486 months between October 1976 and March 2017 –— 41 years –— Berkshire Hathaway’s Class A stock earned an average excess return of 18.6% per year above U.S. Tbills. Annualized volatility was 23.5%. Sharpe ratio: 0.79. Berkshire’s Sharpe ratio of (0.79) is roughly 1.6x times the broad U.S. stock market’s Sharpe ratio of 0.49 over the same period. Among all large-cap U.S. stocks and mutual funds with 30-plus-year continuous track records, those are unmatched numbers. A dollar invested in Berkshire on October 31, 1976, was worth more than $3,685 by March 31, 2017. A dollar invested in the S&P 500 with dividends reinvested over the same period was worth approximately $76. Buffett beat a passive index by a multiple of 48. But he didn’t do it with stock picking! Three researchers at AQR Capital Management –— Andrea Frazzini, David Kabiller, and Lasse Heje Pedersen –— dissected Berkshire’s 50 years of investments through 2013. They expanded and republished their findings in 2018 in the Financial Analysts Journal, which is the most highly respected industry financial journal. Their work won the Graham and Dodd Award for the best published paper of the year. The paper is called Buffett’s Alpha. They found, after accounting for cheap leverage (from the insurance float) and exposure to a handful of publicly documented factor premiums, Buffett’s investment skill –— the portion of his returns that cannot be explained by any mechanical strategy –— is 0.3% per year. That's statistically indistinguishable from zero. In other words, the alpha that Berkshire enjoyed for 50 years (as it compounded capital at 24% a year!) wasn’t due to Buffett’s stock picking. So, how did he do it? He did it by gaining access to a huge amount of investment capital that he did not own, for free. Buffett’s track record was built on leverage. That’s a dirty word for most investors, but it's the secret behind Berkshire. The AQR researchers had access to something most Buffett commentators do not: 40 years of Berkshire’s audited financial statements and the full quarterly history of the public 13F stock portfolio. The researchers asked a specific question: If I take Berkshire’s monthly stock returns from October 1976 through March 2017, and I run a linear regression against a set of well-documented risk factors –— market beta, size, value, momentum, and two newer factors called Betting-Against-Beta and Quality-Minus-Junk (detailed below) –— how much of Buffett’s performance can the factors explain? And after the factors have been stripped out, how much excess return remains? The data show clearly there are a few qualities that drove Berkshire’s results. First, Buffett has always preferred large-cap stocks, contrary to the popular image of him as a small-cap value investor. He buys elephants. Second, no surprise, Buffett buys cheap. Berkshire is almost six standard deviations away from neutral on the value axis. So far the picture is ordinary. Every large- cap value manager in America loads positively on size and on value. Buffett’s genius lies in the last two factors. These last two factors are a little complicated, but please stick with me. There’s a new factor, that, like value and size, characterizes Buffett’s strategy. It’s called Betting-Against-Beta (“BAB”). What it means is intentionally investing in stocks with very low volatility. The BAB factor captures the excess return that accrues to investors who own low-beta stocks. Low-beta stocks have historically earned higher risk-adjusted returns than high-beta stocks. Financial theory teaches that higher beta (higher risk) should mean higher return. But it doesn’t. The opposite occurs, in fact. And Buffett was one of the very first people to figure this out. Why does this factor persist? In an efficient market, once that factor is known to investors, then they should bid the price up on low- beta stocks until it no longer provides an edge. The explanation, per the theory of AQR’s Frazzini and Pedersen’s theory, is that because ordinary investors do not use leverage and seek high returns, they create persistent excess demand for more volatile stocks. (Having worked with retail investors for 30 years, I can assure you that is true.) But, an investor with access to cheap leverage –— Warren Buffett, for instance –— can exploit the mispricing by owning the low-beta names and levering them up to produce market-beating returns. And the last factor that matters to Buffett is quality. Buffett buys companies with high returns on invested capital. Quality-Minus-Junk (“QMJ”) is a factor described by Cliff Asness, also at AQR with Frazzini, and Pedersen, in a 2019 paper in Review of Accounting Studies. The QMJ factor captures the return to owning stocks of high-quality companies –— profitable, growing, safe, with high payout ratios –— against stocks lacking those characteristics. QMJ has been positive and statistically significant in every major developed equity market for which it has been measured. Berkshire’s loading is 0.37, with a t-statistic of 4.6. –– meaning it is highly significant to Berkshire’s results. In plain English: Buffett only buys large, high- quality, low-volatility stocks of the highest quality. But, Berkshire’s results were not, in any way, unusual. Any investor buying these same kinds of stocks would have earned those same returns –– about 16% a year over time. So how did Berkshire compound at 23% a year? To figure that out, AQR’s researchers built a Berkshire replica. They constructed a simple, rules-based, publicly investable portfolio that mechanically tilts toward large-cap, cheap, low-beta, high-quality stocks, and levers it 1.6- to- 1 to match Berkshire’s insurance float leverage. The correlation between their replica’s returns and Berkshire’s were virtually identical. The authors’ conclusion is unambiguous. “In summary, we find that Buffett has developed a unique access to leverage that he has invested in safe, high-quality, cheap stocks and that these key characteristics can largely explain his impressive performance.” Berkshire’s cost of insurance float has averaged almost three percentage points below the Treasury bill rate across 50fifty years of data. In roughly two-thirds of all years, Berkshire has been paid to hold other people’s money. That is not an investment strategy. That is a financing miracle. It is also the living, breathing heart of Berkshire Hathaway. It’s what Buffett built, starting in 1967 when he paid $8.6 million for National Indemnity’s $19.4 million of float. And it is the factor every retail investor admiring Berkshire’s returns has never paid any attention to. The 1.6-to-1 leverage that AQR measured over the full period, financed at this negative cost, explains the dollar magnitude of Berkshire’s returns. How do we know? An unleveraged version of the same stock portfolio –— which you can approximate by looking at the 13F holdings alone –— has earned an average excess return of 12% percent per year. It’s Berkshire’s leverage that magnifies this excess return to 18.6 %percent. How does this square with Berkshire’s reported gains? Berkshire’s 18.6% excess return, plus the T-bill rate that averaged roughly 4.7% over 1976–2017, gives you a total nominal return of roughly 23% per year, which is the figure you usually see quoted for Berkshire’s historical performance. The 23% tells you what Berkshire returned. The 18.6% tells you how much of that return was compensation for taking investment risk, as opposed to the baseline yield every lender to the U.S. government was earning anyway. With both of Berkshire’s “edges” –— systematic factor exposures to cheap, high-quality, low-volatility stocks and roughly 1.6-to-1 leverage delivered with insurance float –— you get Berkshire Hathaway’s 23% annual gains over 60 years. It’s the structure that’s genius, not the stock picking. And that's very important because it means the original Berkshire formula can work for any investor. I show you exactly how, in my new book.

I ran across this video a few days ago and couldn’t stop watching it. It’s about something ordinary & boring, a plastic gas lighter. But it changes how one thinks about manufacturing. That lighter in so many of our homes, holds pressurised gas. It has over 30 microscopic parts, has to pass international safety codes, & travel 10,000 miles by sea, & the total cost of doing all that, materials, labour, freight, every middleman along the way, comes to fifteen U.S cents. So how does anyone make money on this? Turns out almost the entire world’s supply comes from one place: a county called Shaodong, in China’s Hunan province. It wasn’t always there. But today, Shaodong has 114 lighter-related companies packed into the place & between them they source more than 200 different components from each other, all within a 20-kilometre radius. They supply something like seventy percent of the world’s disposable lighters. And the industry alone employs over 80,000 people locally. Nobody there is winning on cheap labour anymore. They’re winning by shaving a thousandth of a cent off the thickness of a plastic wall, or redesigning a base so a few thousand more units fit into the same shipping container. It took my thoughts back to an old professor of mine, Michael Porter. His 1980 book, Competitive Strategy, is still the 1st book most MBAs read, the one that gave the world the Five Forces and basically invented modern strategic thinking. But there’s a quieter piece of his work, on industrial clusters, that never got nearly the same attention, and it is the one that explains exactly what is happening in Shaodong. His argument was that nations and regions rarely win because of cheap inputs. They win when rival firms and specialist suppliers crowd into the same small geography for long enough that they keep pushing each other past what any one of them could manage alone. He found it in the Swiss watchmaking towns of the Jura, in the German printing press industry and in Italy’s ceramic tile and footwear districts (interestingly, it’s the SAME blueprint which built Morbi, in Gujarat, into the world’s second-largest ceramic cluster, now outproducing Italy by volume. I have posted before, about Morbi) None of these started out as giants. The neighbourhood made them giants. Which is exactly why it’s so relevant to India’s climb up the global manufacturing table I’ve also attached a slide with this post that I saw recently and which shows us breaking into the top 5 manufacturing globally. (A quick reference check told me that we may not have overtaken Korea yet, but the trajectory’s clear) That climb has happened on the back of scale: bigger plants, bigger parks, more FDI. I should declare an interest here, because the Mahindra Group set up 2 of India’s first integrated, plug-and-play business cities, in Chennai in 2002 & Jaipur in 2006. Both have been extremely successful. Chennai’s business zone alone today employs 45,000 people.. But I admit that we need to think differently. A park brings in investors and hands them a ready plot, power, water & roads A cluster is a completely different animal: hundreds of small, specialised suppliers, each obsessed with doing a tiny thing better than anyone else, feeding off each other’s presence for years until no outsider can compete with the whole. I think that’s the work ahead of us now. Not just more factories, and not just more parks. Policymakers & developers like us need to start consciously pulling as many of the inputs and resources a sector needs, the toolmakers, the component suppliers, the testing labs, the logistics specialists, into the same neighbourhood. Shaodong and Morbi both got there by accident, one town stumbling onto a way to shave a thousandth of a cent off a lighter wall, the other discovering it had the clay and, later, the gas pipeline for tiles. We don’t have the luxury of waiting for accidents anymore. We need to do it on purpose

I am the Senior Vice President of Workforce Architecture at Cloudflare and I need to tell you about the best decision this company has ever made. We posted $639.8 million in quarterly revenue. 34% year-over-year growth. Record net retention. The strongest quarter since IPO. And then we fired 1,100 people. Not because of the quarter. During the quarter. I need you to understand the sequence because the sequence is the whole point. My team built the model that made this possible. We call it CIRRUS: "Capacity-Indexed Reduction and Reallocation for Upside Scaling". CIRRUS took our revenue trajectory, our margin targets, and our board's stated appetite for what they called "structural boldness," and it determined that the optimal time to execute a 20% headcount reduction is at the exact moment of peak financial performance. Not during a downturn. Not during a miss. During a beat. The logic is simple. When revenue is surging, the market reads a cost reduction as discipline. When revenue is falling, the market reads the same reduction as panic. Same action. Same 1,100 people. Completely different stock reaction. CIRRUS identified a seven-day window where the earnings momentum and the layoff announcement would compound rather than cancel. I found the math beautiful. I still do. We deactivated 1,100 badges between 9:00 and 9:04 AM Pacific on a Monday. People Analytics determined this was the four-minute window of lowest Slack activity. We called it a "clean cutover." Someone in Infrastructure suggested "zero-downtime deprecation" but Legal thought it sounded too much like a product feature. I thought it sounded exactly like a product feature, which is why I liked it. But I deferred to Legal. I always defer to Legal. That is one of the things that makes me good at this job. The people we cut were not underperformers. I want to be very clear about that because clarity is a Cloudflare value. Sixty-two percent had received exceeds-expectations in their most recent review cycle. Fourteen had been promoted in Q3. One engineer in our Austin office — I'll call him Marcus, though that is not his name and the reason I'm not using his name is not that I've forgotten it — had shipped the caching optimization that directly contributed to $14 million in new enterprise contracts. His manager nominated him for the Raygun Award, which is our internal recognition for outsized impact, six days before I added him to the CIRRUS list. He won the award on Wednesday. His access was revoked the following Monday. The ceremony and the termination were planned by different teams in the same building and neither team knew about the other. I don't think this is ironic. I think this is how large organizations work. The left hand builds. The right hand optimizes. Both hands are attached to the same body and the body is performing well. We let Marcus keep the trophy. It's a small acrylic prism etched with a lightning bolt. It costs us about eleven dollars. His annual cost-to-company was $312,000. CIRRUS selected the 1,100 based on three variables. I'm going to share them because I believe in the methodology. First: salary band. Employees in bands 6 through 8 offered the highest savings-to-replacement-risk ratio. Second: visa dependency. Employees on sponsored visas have a 60-day window to find new employment or begin departure proceedings. This creates what CIRRUS categorizes as "low-friction separation" — the compliance timeline is externally enforced, which reduces our administrative burden. I presented this variable to HR and they requested I rename it from "visa dependency" to "mobility factor" in all future documentation. I agreed. The math didn't change. Third: managerial tenure. Employees whose direct manager had been at the company less than eighteen months were 73% less likely to generate a negative Glassdoor review, because the manager-employee bond hadn't fully formed. CIRRUS weighted this at 15% of the selection score. We call it the "attachment coefficient." We told the market the layoffs were an AI workforce pivot. We said artificial intelligence was making certain roles redundant. We said we were reallocating resources toward our AI gateway products. This was a communications strategy. Not a workforce strategy. The AI framing was my team's recommendation and I'm proud of it because it worked. Two analysts upgraded us the same week. The stock moved 8% in five sessions. The entire AI narrative was four paragraphs in a press release that took my comms partner and me an afternoon to write. Four paragraphs. 1,100 people. 8%. I don't know what the per-paragraph return on that is but I think about it sometimes. The actual AI initiative employs thirty-seven people. We cut 1,100 to fund 37. The ratio is not in any of our public materials. There is a Slack channel called #bright-futures# that our Head of People Experience created for the remaining employees. It posts an automated message every morning at 8:45 AM: "You are the ones we chose to keep." The message includes a rotating motivational quote. Last Tuesday it was a Winston Churchill quote about perseverance. The channel has a custom emoji called :survivor: that the Culture team designed. It's a small cartoon phoenix. Nine hundred people have used it unironically. I find this genuinely moving. I think it shows resilience. My wife says it shows something else but she works in education and I think the frameworks are different. The severance was calculated using a model we licensed from the same consulting firm that built our customer pricing tiers. Median payout: eleven weeks. We benchmarked against industry and landed at the 50th percentile exactly, which our CHRO described as "fair by design." The 1,100 will burn through their severance while our stock price digests a 20% cost reduction applied to a revenue base that was already growing 34%. By the time the last check clears, the savings will have funded the first full quarter of the AI initiative. The one with thirty-seven people. My performance review is next month. I've been told informally that I'm on the COO track. The criteria include "demonstrated ability to execute at scale with minimal organizational disruption." The 1,100 people are the execution. The stock price is the scale. The four-minute badge window is the minimal disruption. I meet all three criteria. I designed all three criteria. Not the review criteria. The outcomes. I keep the CIRRUS model on my laptop in a folder called "Workforce Planning FY26." It sits next to a subfolder called "Offsite Photos — Maui" from the leadership retreat we took in January, where we set the annual targets that the 1,100 people spent four months hitting before we terminated them for hitting them. Marcus's desk in Austin has been reassigned. I don't know to whom. The acrylic prism is probably in a box somewhere. Or maybe whoever cleaned out the desk kept it. It catches the light nicely. I noticed that once, when I visited the Austin office to present the CIRRUS methodology to the regional leadership team. They gave me a standing ovation. The prism was on a desk near the back of the room, refracting a small rainbow onto the wall behind me. I didn't mention it. I stayed on my slides. I'm proud of the work we've done here. I think when people look back at this quarter, they'll see it as the moment Cloudflare became a different kind of company. I think they'll be right. I think the 1,100 people would agree, if you explained the math to them carefully enough.