No One Knows How to Make a Pencil

Lex Fridman opens with Urban's Wait But Why post on the big and the small, from the observable universe down to the atom, and asks which world feels more mysterious or beautiful. Urban's answer is not the romantic one. The very small, he says, seems a lot more mysterious; we kind of understand the big, at least inside the Hubble sphere, whereas at the smallest scales we really have no idea what is going on. The big, though, is sexier. He cannot get enough of the bigness of space and the farness of stars.

Fridman pushes on a mailbag question Urban once answered: are we bigger than we are small, or smaller than we are big? Urban runs a clean accounting. The observable universe is about 10 to the 26th meters across. A Planck length is 10 to the negative 35th. Humans sit conveniently near 10 to the zero meters. An atom is around 10 to the negative 15th or 16th; go the same number of orders of magnitude the other way and you land among nebulas, smaller than a galaxy and bigger than the largest star.

If you stick to real-world objects we actually know a lot about, Urban would say we are smaller than we are big. Drop to the Planck length, or to strings, and we quickly become bigger than we are small. Either way we are middle-ish. The detail that makes the microscopic feel violent is a size analogy: if a neutrino were human-sized, a proton would be the Sun. The small, he says, gets crazy small very quickly.

They linger on a related idea Urban likes: an emergence elevator. An ant is one floor; the colony is a couple of floors up, and the colony is closer to being the animal. Humans ride the elevator in both directions. Sometimes we compete as individuals; sometimes, in a protest or a sporting event, we chant and move as one. Fridman wonders whether the feeling of being an individual is just the floor we woke up on, and Urban does not dismiss it. The conversation is already circling the theme that will run the rest of the episode: collective intelligence, not the lone genius, is the thing that actually builds the world.

Pressed for a guess with a gun to his head, Urban says the universe is teeming with life. He cites a paper by Anders Sandberg, Eric Drexler, and colleagues that ran a Monte Carlo simulation over randomized Drake-equation multipliers. The mean they landed on, he recalls, was 27 million intelligent civilizations in the Milky Way alone. Outside the galaxy that mean becomes trillions.

The interesting part is the long tail. The authors argued that some Drake multipliers, especially the probability that life starts at all, are given ranges that are far too tight. Stretch that variable across something like 200 orders of magnitude and the distribution grows a fat tail. Urban remembers roughly a third or a quarter of outcomes in which we are the only intelligent life in the galaxy, and a non-zero slice in which Earth is the only life in the observable universe.

That is the uncomfortable heart of the Fermi paradox. Find microbes on one grain of sand and you do not conclude the beach is barren; survey the whole beach and find nothing, and both intuitions still have a case. Urban will not pick a side. If this were a freak event, he says, the species that happened to be the freak would still think there must be lots of others, and it would be wrong. A sample size of one makes ordinary intuition unreliable.

Fridman names two reasons the silence still matters. The Great Filter is either behind us or ahead of us, which is a practical question about survival. The sadder possibility, for him, is that someone has been trying to contact Earth all along and humans are too dumb to notice — or that the addressee is Earth itself, a planetary intelligence of which humans are only a fingertip. Urban grants the Copernican pattern: science keeps yanking humans off the cosmic throne. Our understanding, he says, is still very young. Humans may be a curious little side effect of something much bigger.

Urban reaches for a classic demonstration of collective intelligence: there is not a single human on Earth who knows how to make a pencil. The wood, the chemistry of the yellow paint, the eraser, the mined metal, the graphite — no one person can collect and assemble all of it. Together, a pencil is child's play. That gap between individual competence and civilizational competence is, for him, the real magic of being alive now.

Then he offers a meaner thought experiment. A dickish witch erases every invented material thing at once. Everyone is standing naked on Stone Age Earth, but every brain is intact and every person has a note explaining the deal. Produce one working iPhone 13 that would pass in today's Apple Store, and she restores everything. How long? Fridman is honest: a random hundred or thousand people are screwed. Urban adds the ugly details. There are no cranes, no factories, no grid. Everyone is wherever they were, so Manhattan is a grassy island. Even polymath engineers cannot mine without tools they do not have.

If everyone knew the mission and a handful of leaders could coordinate, it might be possible, at the cost of a very long time. Urban expects the first thirty days of hope to decay into new governments, sabotage, and mass starvation. Modern people, he notes, do not know how to farm the way hunter-gatherers did, and they have an ethic that treats murder as worse than their ancestors did, which does not help them survive. The witch's bet is not about phones. It is about how much of what we enjoy is a product of collective intelligence, and how brittle that intelligence is.

Fridman, who arrived from the Soviet Union at thirteen and spent months eating bananas because they were suddenly abundant, uses the supermarket as the same lesson in reverse. Avocados and bananas are cartoon fruit compared with their wild ancestors; crackers and pasta would have stunned hunter-gatherers. The darker question is resilience. If possessions stay and people start disappearing, how many can vanish before the system is lost? Urban reaches for Elon Musk's figure of a million people to make Mars self-sustaining — about one seven-thousandth of Earth's population, and a specialized million at that. Kill half of Earth at random and the species should be fine, he thinks, though it would be an awful tragedy. Kill three quarters and there would be a rough patch, but he is almost sure the species would still hold. It depends who you kill. Knock out the electricians, mining engineers, and factory hands, and the network can collapse.

Musk's million is not a random crowd. Multiplanetary, as Urban uses the word, is often misused. Armstrong on the Moon was a great achievement for mankind, not a great leap for life. If the first fish onto land had given the shore a high-five and gone back into the water, that would have been a statue-worthy feat for the fish, not a leap for life. The leap happens when the change is permanent.

The same test applies to Mars. Only when ships from Earth could stop coming — because of a catastrophic World War III, say — and the people there could still turn their numbers into billions, build ships, and recolonize Earth, would we officially be multiplanetary. That is why the million has a specific skill structure. It has to be able to rebuild computers, mines, farms, electricity, plumbing, factories, and transport. Someone, Urban jokes, would still run a YouTube channel.

What excites him about Martian politics is the blank canvas. The 1500s had unclaimed land to seize, name, and experiment on. Earth is now so optimized that Crimea switching hands can jolt the world for years. Mars is territory, naming rights, and maybe new governments. The United States could found a modern democracy in the late 1700s partly because it had a clean slate; imagination is less often the bottleneck than the missing canvas. Urban hopes a colony would have sovereignty from day one. He also admits it could get weird: geopolitical tension with Earth, a new colonial race, us-versus-them tribalism, even war over space territory.

In 2018, drunk, he bet a friend $10,000 that a human would set foot on Mars by the end of 2030. He was more confident then; now he thinks it will be around that date, or a little later. His friend said it would not happen in their lifetimes. SpaceX, Urban argues, is almost there technically and is delayed by safety: cargo ships, looping flights, the Moon in the way, and a 26-month synodic window. Even 2024, an early Musk estimate he had heard, will not happen. 2027 and 2029 are his remaining shots. SpaceX exists, as he hears Musk tell it, for two reasons: life insurance for the species, and the greatest adventure in history. The premium on that insurance, stacked against extinction, looks cheap.

Asked what makes Musk special, Urban lists intelligence, motives he believes are genuine, ambition, and work ethic, then sets them aside. Plenty of people have those. The unusual quality, he says, is that Musk is sane in a way that almost every human is crazy. We are programmed to trust conventional wisdom over our own reasoning, for good Pleistocene reasons. Do not eat that berry; tie the spearhead this way. That knowledge was a mini collective superintelligence, the ancestral version of not knowing how to make a pencil. Disagreeing with the mountain god got you ejected from the tribe.

The setting fails in a rapidly changing world. Urban pictures an old bucket that used to sit still and collect water; today the bucket is moving, so wisdom does not accumulate, but our brains have no "fast-moving bucket" setting. His grandmother's advice on love and living near family is still gold. Her advice to go to law school for security is software written for a world that is no longer here. Conventional wisdom, he thinks, is often ten or twenty years out of date. Most of us still lack the guts to act on our own reasoning when it conflicts with what everyone else thinks.

Zip2 was internet advertising when people scoffed. X.com, which became PayPal, asked people to put bank information on the internet when that sounded fake. A friend compiled a montage of rockets exploding to talk Musk out of SpaceX. Landing rockets was the classic analogy trap: if it could be done, NASA would have done it, or the Soviets in the 1960s. In each case Urban thinks Musk gets too much credit for Einstein intelligence or seeing the future. The guts were to look at present reality, assess probabilities, ignore noise, and trust the math on the envelope.

The name for that move is reasoning from first principles: start from axioms, build a hypothesis, test it. The shortcut is reasoning by analogy, photocopying other people's conclusions. Analogy is usually good; you do not want to reinvent the wheel every morning. It is a bad tool for the handful of choices you will score on your deathbed — whom to marry, where to settle, how to raise and educate children, which career turns to take. Great artists and founders have a moment of trusting their own reasoning: Airbnb was turned down by more than sixty VCs. First principles often fail, Urban warns. You walk into fog without a map. When they work, everyone flocks over, as they did after the iPhone, which did not start from a cooler keyboard on a flip phone. It started from what a mobile device should be.

Urban had just tweeted that he wrote the last word of a book on why society feels like such a mess. The top reply, which he says he earned as a known procrastinator, was that now he just had to work on all the ones in the middle. The puzzle that pulled him in is a split in the graphs. Max Roser's charts show almost every important metric getting better. Meanwhile the share of Americans who hate other Americans, hate their country, and hate the president is on a steep rise. New people who are simply jerks were not born. Something in the environment changed.

His kindergarten model is that behavior equals human nature times environment. Nature is nearly constant; environment has been moving fast; behavior is the output. Decades ago a president might be liked by 70 or 80 percent of their own party and around 50 percent of the other. Now he cites numbers more like 40 percent own-party and 10 percent the other. Presidents did not suddenly get uniquely worse. On one racism metric — how upset you would be if your child married someone of another race — the number has been plummeting, while racial grievance has been skyrocketing. Same society, opposite arrows.

Urban calls the psychological trap political Disney World. In those movies Scar is totally bad and Mufasa is totally good; you do not see the hero's flaws or the villain's upbringing. Digital brains slide toward ones and zeros. No one is a one and no one is a zero. We are all 0.5s, or maybe 0.4s and 0.6s: worthy of criticism and of compassion. A montage of anyone's worst moments would make them look awful; a montage of their best would make them look like a god. What makes him hopeful is that a bunch of 0.5s is good enough to build a decent society, and that there is more love in humans than hate.

He still remembers Derek Jeter's last game at Fenway. As a Red Sox fan who hated the Yankee most of all, Urban nearly cried during the standing ovation. It felt like a little fist-pound: of course we actually love each other. If the problem were human nature, he would be more despairing. The independent variable is the environment, which can change. He pictures the species in a five-million-mile canoe, each person with a paddle, a fork somewhere ahead, utopia on one side and dystopia on the other. Tech is exploding hard enough that the future is probably not going to be merely okay. Squabbling into a civil war would make all the futurism irrelevant.

Urban had tweeted a prompt about changing one's mind. Musk replied "brain transplants"; Sam Harris replied "nuclear power." Urban's own view of nuclear, he admits, is still reasoning by analogy. A friend, Isabel Boemeke, is a nuclear-power influencer, and a cluster of smart people he respects have dug in and come back saying it is an obviously good, emissions-free option whose meltdown and waste fears are overblown. He has not done his own deep dive, but he is a strong advocate on that secondary evidence. What strikes him as odd is how many people who fear climate change treat nuclear as an ideological mismatch rather than as a rational response to that fear.

On climate itself he is again analogizing: it looks like a true existential threat, and he feels hopeful, but he has grown uncomfortable with fear-mongering. Too many people, he says, are interested in manipulating the public with fear, and one side effect is that it grows skepticism of science itself. He has not seen enough calm, first-principles modeling of what catastrophic effects actually look like in ten, twenty, or thirty years — sea-level models, migration, resource constraints, how societies respond. The pandemic, for both men, was a warning about large collective decisions. More of those are coming, including the near-term possibility of parents selecting embryos for IQ, first available to the wealthy. That is why Urban wrote the book: the species needs its wits.

His other simple model is a higher mind and a primitive mind. The higher mind treats ideas as disposable hypotheses. Criticism is no big deal. When higher minds team up, they throw ideas into a room and shoot arrows at them; true ideas bounce the arrows and rise, false ones get shot down. That is a good scientific institution, and it is how a bunch of flawed, medium scientists create a superintelligence. The primitive mind lives tens of thousands of years ago. It does not treat ideas as separate objects. It holds some of them sacred and identifies with them. When those minds team up they build an echo chamber. Papers get retracted, people get fired, and no arrows can be fired. The collective becomes dumber than a single reasoning individual.

Climate change, he says, should have been a higher-mind problem: complex, long-term, full of nuance. Instead it has been sucked into the red-versus-blue whirlpool. Offer a nuanced take and you may be branded a denier, which is a social penalty for firing an arrow. The big American brain, he says, is drunk at the wheel, heading into a future of gene editing, AI, and other high-stakes choices. The opposite of an echo chamber, a term he coined because he thinks we need it, is an idea lab: people toss ideas, disagreement is the game, nobody is offended, mutual respect is assumed. Echo chambers have lately found ways to expand into places with a strong immune system, including universities and science journals. If you grew up in an echo chamber, an idea lab feels like an insult. If you grew up in an idea lab, an echo chamber will kick you out and brand you.

A mailbag question asks why we prefer a film we have not seen, but will listen to a song hundreds of times. Urban brings the two minds back. The higher mind is taking in information; once it has the plot, rewatching has little point, which is why few people play a podcast twelve times for the facts. Music goes into the primitive mind. The first listen may be merely catchy, or even annoying; by the tenth you love it, and by the fortieth a song you hated on the radio can feel like yours. The sound carves a pathway. Your brain is dancing with the music and knows the steps. A first listen is an awkward dancer. A movie you truly love is the exception that proves the rule: you will watch it twelve times because some of it has become music. Fridman adds that people who replay a podcast are usually in love with the chemistry, the beat of a personality, not the information.

That split is a warmup for Neuralink, the subject of Urban's wizard-hat essay, which Fridman says changed how a lot of people think. Urban calls it Musk's most ambitious project, more than colonizing Mars. Mars is a bunch of people going somewhere far. Neuralink is changing what a person is. Near-term medical goals — paralysis, stroke, Parkinson's — would already be huge. The grand goal is augmentation: a brain-computer interface in a healthy person that confers superpowers, cheap enough, eventually, to be more like LASIK than like a rare surgery.

An early demo he likes is music without speakers. A song is a speaker vibrating air, vibrating an eardrum, vibrating bones and fluid, stimulating auditory-cortex neurons. If enough Neuralink electrodes got good at stimulating those neurons, you could play a song in a silent room that only you hear. Spotify would get a new output: speaker, computer, phone, headphones, or brain. The more radical step is thinking together. Musk's example: you leave a horror movie and say it was terrifying, having chosen among five buckets — horrifying, terrifying, scary, eerie, creepy — while your actual experience was far more nuanced. Language, Urban says, is very lossy. You send a low-resolution package; the other person guesses at the high-res original.

With a high-bandwidth link, two brains could collaborate without words, internally or externally. It would be a skill; kids would be good at it and old people bad. Eight designers could beam a dress from a mental image onto a screen. If it works at that level, Urban thinks it becomes a new common-era line: anyone who lived before widespread brain-machine interfaces will look prehistoric. Fridman is sad about losing physical presence. Urban's first-principles reply is that all the components of intimacy can be engineered in cyberspace once you are stimulating visual cortex rather than wearing a headset. Touch and pheromones may be last to go. A masseuse could work on a dummy body while a thousand people feel it, the way a book or a Taylor Swift recording already scales one performance.

Fridman invites Urban onto AI, including safety, Autopilot, Boston Dynamics, Roombas, and recommender systems. Urban's near-term excitement is more intimate than alignment essays. Future wearables, he thinks, will make a Fitbit look like a hilarious bracelet. Sensors in blood or infrared on the skin will collect a hundred times more data: a full genome we actually know how to read, real-time blood chemistry, the exact width of heart arteries, traces of every virus that ever visited, heart rate, muscle mass, weight, maybe even emotions as chemistry. Lisa Feldman Barrett's work, he notes, already warns that a popup asking "how do you feel?" is a bad instrument.

He wants the model to tell him what to do, hooked to a meal service with a million ingredients, supplements, and vitamins. He would give it broad goals: gain muscle, keep weight, have more energy, avoid cancer, age slower. A prompt would say he should eat in fifteen minutes; a drone would drop a meal through a slot in the wall, assembled for that moment's genome, blood, mood, and calendar. It might offer coffee only inside a fifteen-minute window, because it knows how long caffeine stays in his system and when he has to wake up. Most people, he says, will hear this and call it dystopian. He thinks it sounds incredible. Nobody would look back and miss making awful daily choices.

Those, he argues, were never the important decisions. Attention and willpower should go to children, work, and other people. One-size-fits-all health advice — a keto article written for the average human — will look primitive next to a model whose advice is for you. Fridman notes that in a capitalist society the same data is more likely to compete to deliver a delicious healthy meal than to feed an authoritarian fantasy, which is why Urban pivots to free speech. Free speech is the mechanism by which the big brain thinks straight. The First Amendment is one ingredient; a culture of free speech is the other. If culture destroys your life for a nuanced take on police shootings, schools, or climate, the amendment is still there and the culture is gone. In a vigorous culture you can say something dumb, people laugh and correct you, and nobody decides they have seen your "real colors."

Urban once tweeted that thirty minutes of reading a day equals a thousand books in fifty years. He likes the compiling metaphor. Legendary people had the same number of days; they put one brick in a day until they had a monument. Two aspiring writers can live 98 percent identical weeks. One writes two pages every Sunday morning and, at that clip, produces a book every two years — ten books in twenty years, a real career. The other remains a wannabe who talks about how they used to have it. The gap looks magical from the outside. It is a small, repeated act.

Urban is a slow visual reader and was frustrated by it. Audiobooks changed the math. He throws one on in the shower, while brushing his teeth, making breakfast, dealing with the dogs, until he sits down, and at that clip he can finish a book a week or every ten days. Fridman, who listens to dense history, treats attention as a skill. The ten-second back button is his tool; he will tap it three times. Sometimes a sentence is so good he needs to pause. Sometimes he is thinking about something else, which means his mind wants to work on another problem, so he pauses the book rather than pretending to listen. Urban adds that related mind-wandering is not wasted: the brain is categorizing what it just heard. Continuous drift is the signal to stop.

Fridman then describes Anki, a spaced-repetition app medical students use for tens of thousands of facts. You dump in a line you want to remember; the software brings it back sooner if you fail and in a month if you succeed. It offloads the executive job of deciding what to review. The interface is ghetto, like Craigslist, which is why a hardcore community defends it: they want a text box, not features. Urban cannot believe he had not come across it. He already collects factoids in a document for readers he thinks of as friends; Anki would let a cool detail surface later instead of disappearing. Fridman guarantees he will probably write a blog post about it.

Fridman asks how to avoid procrastination, putting "suffer" in quotes. Urban takes the quotes off. Treating it as a silly, cute problem is part of why it ruins lives. The writer who respects the part of the brain that decided to do two pages on Sunday will do the pages. The procrastinator wants it just as much, maybe more, and leaves it on a list to collect dust. The loss is the career, the book tour, the life not lived — changing a career, leaving a marriage, launching a startup, firing someone who must be fired, going out to fundraise. Those things have no exam date. Short-term deadline panic is a different animal. The Panic Monster can still get most people to show up for a test. It cannot make them start the life they know they should be living.

He splits fixes into duct tape and repairing the boat. Duct tape is an external Panic Monster: book a venue two months out, tell forty friends you will play songs, or promise a painful donation if a task slips a week. That is not a repaired boat; the incentives were already selfishly aligned and something still got in the way. The obvious writing boat, for him, is waking around eight, coffee, and a sacred nine-to-noon block. Most great writers wrote in the morning. Five weekdays of three focused hours is fifteen hours, a big week, a B every day, and enough compiling to hit every writing goal he has. Instead he revenge-procrastinates late, opens email, takes a three-hour call, or hits a wall and leaves because the block was never sacred.

His current tape is Alicia, Wait But Why's one employee. From nine to noon she sits where she can see his screen and does her own work. That is enough. Procrastinators are not usually kicking and screaming; they are standing at the edge of a cold pool all day to avoid one second of entry, then loving the water once they jump. At the gym with a trainer he finishes a full workout in 45 minutes; alone he does a set and phones for ten. He leaves for the airport at 3:36 when 3:00 would have been calm, then immediately wonders why he did not just get in the car. Alicia is the person in the room who supplies the "ugh, class is starting" nudge he does not generate internally.

The famous characters are the Instant Gratification Monkey and the Rational Decision Maker; for a procrastinator the monkey wins, except on habits like brushing teeth where the monkey does not even see an option. The Dark Playground is the toxic middle: you are doing something that is fun on paper while you should be working, so it never recharges you. Urban spent five years on a book that needed two or three, over-researching, stuffing in facts he would cut, dropping mystery dates with his wife, waking thrilled and going to bed in regret. He has not fixed the boat. He has some good duct tape. The goal is black-and-white time: a few ferocious hours, then clean-conscience play, instead of a gray day of guilt.

Looking forward, Urban has a long list of things he is excited to make and knows he will never finish. Publishing is the joy; doing the work is not. He still wants more long Wait But Why posts, another book after this one if the first two feel good, more videos, and another travel series like the one in which readers voted him through Siberia, Japan, Nigeria, Iraq, and Greenland. He wants to try fiction, science fiction, and a musical he started with Ryan Langer before Wait But Why took off. He likes going on other people's podcasts more than he expects to like interviewing. He cannot help jumping in, which made him a bad panel moderator and, he suspects, would make him a bad interviewer.

Fridman's last question is the meaning of life. Urban wants to stick around as long as he can. If extreme life extension or uploading is on offer, he will take it, because who does not want to see how cool the year 3000 is? He does not want eternity. After writing about Graham's number he is sure nobody who understood eternity would want it. He wants the choice to sign off, complete with a ceremony, when he is truly over it. Fifty centuries sounded about right; he does not know that he wants more. Wanting fifty and getting one, he says, makes one a hideous number.

The real answer is a warm, fuzzy connection. He would not want to be a journalist whose personality hides behind the prose, or even a biographer. If he can get something out of his head and a reader says they thought that too, or that they feel better, or that they feel seen, that is the thing. We are the same four-limbed creatures on one rock in one little moment, suffering in similar ways. If on his deathbed he has had a ton of human connection and made a lot of people feel not alone, that is enough. He does not always feel it while typing. In-person events and emails bring it back. Publishing is the moment the connection happens.

Fridman offers a last scene: Urban's final blog post, written from Mars after a rocket malfunction strands him. Urban laughs. He would like to go for a week and come back. If he could still write, and people on Earth could still read it, that would make it so much better. Being alone with nobody realizing what happened is the scary version. Fridman says it would bring out great writing. Urban thinks so. Fridman closes with Urban's own line: be humbler about what you know, more confident about what's possible, and less afraid of things that don't matter.