Anonymity Is a Tool: How a Yale Freshman Built a Campus Where Students Could Speak

Librex, Schiller tells Fridman at the outset, is an anonymous discussion feed for college campuses: a place where verified members of a school community can talk about ideas that matter without attaching a name. It began at Yale, his educational home though he had not yet graduated, moved to Dartmouth, opened to the rest of the Ivy League during quarantine, and, the week of this recording, to MIT and Stanford. The Ivy League, he recites in one breath, is Harvard, Yale, Princeton, Dartmouth, Columbia, Cornell, Brown, and Penn. He expects Fridman to make an MIT account that same day.

The conversation detours through podcast editing software that can change a speaker's words, a deepfake with a constructive use. Schiller's hope runs the other direction from the cultural panic. If everything can be faked, everyone will have plausible deniability, and people will grow more skeptical, check multiple sources, and actually reason about what is true. Institutions such as *The New York Times* once held a reputation as sources of truth. When you can no longer trust any single source, he says, you start to think on your own.

That lands Fridman back in the Soviet Union, where you could not believe any one news source; you had to think, and you had to talk to friends. Schiller pictures billions of minds as a point cloud, and what a society treats as true as almost a clustering algorithm: a bunch of people believe the same thing, and that becomes true, alongside other truths, in a battle of truths. Maybe the idea of a capital-T truth will lose power. The downside of declaring something true, Fridman adds, is what religious people call scientism: once science has announced a result, you are no longer allowed to question it. Science is a moving mechanism. Truth might need to be renamed as a process, not a destination.

They rewind further. Civilizations clustered around common values, morals, and places, each growing its own truth — Torah, Buddhist texts, many articulations of similar things. Yuval Noah Harari's *Sapiens* pushes the same line back to the thing that made humans special: clusters of ideas held in the mind as stories, passed along, growing. Money is another belief system that has power only because we believe in it, and Bitcoin is the current exhibit. Then Schiller steers back. What, Fridman had asked, was the founding story?

As a freshman, Schiller heard the same complaint from friends: it was hard to raise a hand in class, and hard to be vulnerable outside it. Yale is not that big. Everyone knows someone who knows you. People come there to be themselves and to do the intellectual work the brochure promised, and the culture made both difficult. He took a global-affairs course because the instructor, Professor Wright, was an expert on the Middle East conflict. Near the end of the term he went to office hours, one on one, and asked when they would finally hear that expertise. Hearing an expert go beyond the textbook, then pushing back, was the reason he had come to Yale at all.

Wright's answer, as Schiller recounts it, was blunt. He used to teach the topic every year, including a small seminar around a Harkness table where he laid out his research and students debated it. About two years earlier, a student reported him to the school. He realized his job was at risk. The safest course was not to broach the topic. "You can say whatever you want," Schiller remembers him saying, in essence, "but I'm not going to be part of it." Fridman locates the strangeness in the reporting mechanism itself: a complaint that an idea is painful, disrespectful, or unsafe can outweigh the idea. It does not take a crowd. One or two people can be enough, and he is not sure how an administration fixes that short of firing everybody.

Schiller refuses the simpler story that the Ivy League is broken. He separates faculty from administration. Faculty, he says, are often the boldest thinkers, the people in possession of the deepest truths and the capacity to educate. When critics attack academia, they are usually attacking the administration. Those offices have too many stakeholders, and they often lack two things: courage, the willingness to step into the fire, and grace, the skill of walking through it without being rude, insensitive, or inarticulate. Neither was trained into them. The faculty, within the bounds of a career, can still be some of the bravest people in the building. That gap was the spark.

Schiller wrote articles, talked to friends, and concluded that the fix had to be cultural and bottom-up. The energy was already there. Yale's most recent institutional assessment had students, faculty, and alumni all pointing the administration at the same first request: cultivate more conversations on campus, especially more difficult ones. A Gallup poll found that 61 percent of college students nationally were afraid to say what they believed because of campus climate, and Schiller suspects the Ivy League is worse. Fridman asks him to repeat the number. It is a majority. It is not a conservative fact or a liberal fact. You do not even need a formed opinion. You only need the fear.

Fridman pushes the cost of that fear. Metaphors are how teachers explain things, and caution is death to humor. Comedians can talk without that freeze; professors cannot. Schiller takes the point further. If you are afraid to speak metaphorically, afraid to be imprecise, it becomes difficult to think at all, because messy, sloppy language is the tool humans use before they have mathematical precision. That messy step is the beginning of new knowledge. If students cannot excavate ideas with each other in that ungainly way, he wonders how much they will still be able to learn.

The grassroots version, as Schiller saw it, had to be personal: people meeting people and discovering they were not the only ones who felt alone, or who held an unpopular view. He started coding. He did not have much experience. Fridman, who has spent a life among software people, calls the decision beautiful. Schiller did not hire a technical cofounder. He bought a Mac, because kids live on their phones and he wanted a passing thought in the dining hall to be publishable. He downloaded Xcode, searched YouTube for a Stanford Swift tutorial, and clicked the video with a million views. It was an introductory course. Most of the people in that class, he later realized, were already software engineers or computer scientists. He spent a lot of time on Stack Overflow.

The goal was never a finished product. "I'm trying to prove out an idea," he says. "I'm not trying to make a finished product. I'm just trying to get to the first step." If he kept reaching the next step, he would not die, he would learn, he might meet people, and the world would see he was serious. The MVP was as thin as it sounds: put text on a screen, post, comment, let anyone with a Yale email register. He built a database with no knowledge of databases. Friends in computer science asked why he had stored it in a heap. He did not know what a heap was. He had done it because it worked.

Fridman extracts the lesson engineers keep missing. The rabbit hole of the proper way to do it is how you wake up five years later still on the login page, or at Amazon, having never shipped. Schiller was chasing a different signal. Telling someone an idea is one thing; putting it in their hands and watching their eyes is another. He walked around Yale's cross campus, ran up to people, and asked them to try it. Some were not interested. Some got it immediately. He did not need 50 percent to like it. He needed 5 to 10 percent to love it, and they would tell the next 5 to 10 percent. All of that original code has since been stripped out. Friends told him to stop coding and do the rest. He never intended to be an engineer. He only needed to validate the idea.

Traction at your own school, Schiller says, is one thing. Going somewhere no one knows you and signing up 90 percent of the campus overnight is another. In March 2020 he wanted to know whether Librex could go from one campus to two. He knew nobody at Dartmouth. He sent cold emails asking whether he could sleep on a dorm floor for two days during finals. People told him no one would answer. One person did. He took a train knowing only a first and last name. Dartmouth, in Hanover, New Hampshire, was more remote than he expected. There were no Ubers. He eventually found the one person who could drive him to campus, a ride that took hours, and slept on the floor.

In the morning he asked about printing. Dartmouth's copy room was free. He printed about 2,000 posters with a QR code. The staffer told him, in so many words, that he did not know what the kid was doing but he needed to get out. Schiller then slipped a poster under every dorm door until his legs burned, said hello to everyone on the paths, and worked the cafeteria table by table. Twenty-four hours later he sat on a bus hitting refresh on the account-creation panel as the numbers jumped by hundreds. The pitch he repeated until it became the line he still uses was factual and short: this is Dartmouth's anonymous discussion feed. He had learned he did not need to explain why to try it. People already knew. They had been waiting.

What followed interested him more than the signup curve. Professors wrote poems during finals week. Two women met on the app and started a finance club. Yale's campus press later wrote up two women who met on Librex and began a relationship. People threw parties. Schiller's summary is the one he keeps returning to: when you allow people to be vulnerable and social, they connect, because people have a natural desire to connect. Pair that desire with the freedom to speak, and they connect as themselves, not as a performed image. Those are not the same thing.

The post Schiller remembers most clearly came from a Mexican international student at Yale. It was edgy and still drew hundreds of upvotes inside the Yale community. The writer described La Casa, the Mexican and Latino cultural center, as a place where he did not feel welcome: he was Roman Catholic, he was international, and he did not fit the center's agenda. The house that was supposed to be home made him feel more outcast and more alone than anywhere else on campus. Hearing someone who should have been supported by that culture say it was not helping him, Schiller says, was powerful. In a named environment the sentence is almost unsayable, because it offends the institution, the identity, and the good intentions at once.

Fridman asks the obvious question about anonymity: it frees a voice, and it also frees trolls. Schiller's distinction is careful. Anonymity on Librex can be liberating, a way to talk about politics, relationships, and mental health without a performance. It can also let people say things they would not say with a name attached, which is sometimes useful and sometimes requires moderation. The important sentence is the one he keeps repeating. Anonymity is a tool in Librex. It is not the purpose of Librex. It is how you get authentic conversation given the campus climate. And it is not Reddit anonymity. Everyone is a verified student. You know their school. There is one account per person. People share a vernacular and a set of events, and they feel ownership of the community.

That design shows up in the feature Schiller is proudest of: an anonymous direct message to the original poster of any thread. The recipient sees only a school. If they accept, the two of you are one-on-one and can swap a number, Snapchat, or email and meet in person. A Harvard Law professor told him it was his favorite part of the app. A coin collector found another coin collector that way. There are no usernames, only school names, for two reasons. Safety: these communities are small, and in 2021 a few posts would be enough to identify someone. And ideas: Schiller does not want people boxed by who they were in a previous thread. Fridman, who is not a student, felt like an outsider in the product, which he reads as a feature. Each school is a village, and the Ivy League is a larger public square. Schiller says that coupling is the point. Social products should be bound to a shared lived experience, not to a map.

Fridman had read that one reason Schiller was rejected from Y Combinator in the final rounds was a refusal to sell user data. Schiller does not confirm the YC story. He answers the principle. People on Librex talk about political opinions, mental health, and relationships — "some of the most intimate thoughts that people were having, point blank." Turning those beliefs and secrets into a currency he bought and sold did not seem ethical or moral. "I don't think I could sleep at night if that was what I was doing."

Fridman believes there is a middle path: transparency, user choice about what is shared with advertisers, and deletion that is actually easy. He had just spent a month locked out of Instagram by a bug, someone hammering an API with his account until the system decided he was a bot, and he had to prove he was himself. Disappearing from Facebook, he says, should be possible and in practice is extremely hard. If you want a good product, make leaving effortless. Librex added a button labeled "forget me." Two clicks, and the email is gone from the database. Fridman finds that moving. If he could disappear at any time and still trusted the founders, he would give more, not less. What people want is trust and ease of escape.

Trolls exist even at Yale and Dartmouth. Schiller's answer is a short list of principles, volunteer moderators from every Ivy League school, and a constant conversation about standards. The moderators are not a random intern pool. Professors and students reach out; he looks for people in tune with their campus, from diverse backgrounds, who understand the point of the product, and he walks them through both clear past cases and murky ones. They can take posts down. He can reinstate them. He still answers almost everyone who writes, which he knows is an early-days luxury, but he does not think he can ever fully abdicate moderation. Open discourse is the product. Light-touch, community-guided moderation is how you keep it.

Asked to define freedom of speech inside that product, Schiller starts from an unhelpful fact: he is a math major, and the phrase is a philosophical mess. The name Librex was meant to point at free exchange. His purpose was as much intercommunication of ideas, repugnant or otherwise, as possible, within the law, without harassment or doxxing, focused on ideas rather than people. The practical rule that survived contact with users is simple: no sweeping statements against core identity groups, meaning race and religion, even when the generalization is dressed as a joke. Be specific. Broad claims against groups are usually hackneyed and go nowhere. As the community grows, people will game whatever rules you publish. That creativity is human. Some vagueness in the rules is, uncomfortably, part of the design.

Yik Yak is the obvious ghost. Schiller says he started building without knowing it, then designed around its failures once people mentioned it. The differences he cares about are not a single postmortem slogan. Librex is community-based, not geographic; verification and one account per student let him control who is in the room; expansion is intentional, with moderators who know the school, rather than a spring-break explosion from 12 campuses to a thousand plus high schools; and the posts are long-form, not Twitter-length yaks, which makes room for nuance. A small minority produces most of the worst content, a pattern he has also heard from the founder of 4chan. Mission, he and Fridman agree, changes who self-selects in. Yik Yak, by its own account, was a place to yak. Librex claims a larger purpose, and students have started treating it as faster than the school newspaper.

Fridman asks whether the thing scales past colleges, the way Facebook did. Schiller's unit of community is the school, perhaps a system such as the University of California or a conference, not the whole world at once. Company emails could verify employees the way .edu verifies students, and Google or Microsoft as anonymous villages has crossed his mind, but he understands the college need from the inside and is not pretending to understand Google. The point of growing slowly is to learn from each community, not only to add it.

Fridman is less interested in Parler as a product than in AWS removing it. He had treated cloud infrastructure as a competitor-neutral highway. Schiller is bothered less by the specific takedown than by the pattern: something people rely on, build on, and assume is position-agnostic is becoming ideologically discriminatory. Amazon is a private company and can do what it wants; he supports that right. He is less sure at what point a company that sells storage and compute should stay agnostic and let developers build. Fridman tries a private-highway analogy: if the road announced it would only allow electric vehicles, many people would cheer, and then you would have to think about the slope, the lives disrupted, and the effect on competition. Some layers of existence, Schiller says, should not be messed with. Storage and compute feel like that layer.

Neither man is eager for government to patch it. Fridman would rather see capitalist competitors force AWS to do better. Schiller's heavier point is not that the Parler decision was uniquely wrong. It is the slope, and the fear it installs in people who are trying to build good products. Fear, he says, is why you fail to create the best thing you could create. That is the same logic as anonymity on campus. You do not do your best work when you are afraid.

They turn to Ronald Sullivan, the Harvard Law professor who joined Harvey Weinstein's defense and represented Aaron Hernandez in a double-murder case, because he thinks that is how you test whether you believe in the rule of law. A minority of students protested to have him removed. The argument was safety: it is unsafe to have someone who is, in effect, supporting Weinstein nearby. Schiller's split is clean. Safe spaces are fine in private life, at home, in a religious organization. The problem is an institution encouraging or endorsing them. What harm is being mitigated? Often it is the harm of confronting opinions you disagree with, even opinions you find repugnant. Conflate that with safety, and liberal arts education dies. We still need safe intellectual environments, he says. The current use of "safe space" does not belong in an intellectual context.

Fridman sees an asymmetry of power that lets "safe space" and "hate speech" slide until they mean whatever the speaker needs in order to silence thoughtful people. There is a linguistic project as well as a pragmatic one: to treat words as unsafety, harm, and violence. Once words are violence, and words are criminalized, you are in authoritarian territory. Schiller is a child of Holocaust survivors and pogrom survivors. Fridman comes from the Soviet Union. Both agree that hate and love run through most people's blood from history.

Librex had opened to MIT that same day. Fridman wants someone to start a thread about Jeffrey Epstein. Faculty connections, things said, and money taken had a huge impact, he argues, and the conversation still has not happened. Administration is busy cleaning up, offering boilerplate, pretending nothing is broken: a bad man did bad things, a few professors were a little clueless, everyone look away. Harvard, he adds with an edge, handled it "better," well enough that people almost pretend Harvard had nothing to do with Epstein. The topics underneath — sexual assault, what funding is ethical, whether Defense Department money is different — are exactly the conversations a university should be able to hold.

The name that hurts him personally is Marvin Minsky, a founder of artificial intelligence and one of the reasons Fridman wanted to come to MIT. A mailing-list thread about Minsky and Epstein was shut down quickly. Richard Stallman, who wanted clarity and did not speak with much grace, was pushed out after a few people protested. What bothers Fridman most is that the conversation did not continue or grow. There was complexity and curiosity, and it was interrupted. Minsky's name was dragged through the mud without being resolved. He does not know what to think, and the only way through is to keep talking. That is the same point the episode opened on: truth is conversation.

Schiller's reply is about closure. Librex usage spikes whenever there is a tragedy on campus. A few days earlier there had been a suicide at Yale, and people came to say rest in peace and to talk about the environment that can push someone there. People need a place to speak, especially when no one wants to be the tallest blade of grass. Fridman says that silencing fear, the kind that makes people censor themselves, is one of the things he cares about most.

Before Librex there was chess. Schiller played tournaments until about 13 and reached a 2000 USCF rating. Strong 13-year-olds in local events end up across the board from adults. He played a mid-forties man, lost a game he still remembers, and stayed for a postmortem, a teenager and a 40-year-old hanging over the same position. Even then he could tell the man was brilliant, and not only at chess. He went home, looked him up, and found Bill Chen, a World Series of Poker champion. At 14, at another tournament, he followed Chen into an elevator pretending he needed to go up, just to talk, and suggested changes to a book. Chen said he had been thinking of the same thing. For a 14-year-old, that was enormous.

Chen's *The Mathematics of Poker* pulled him into game theory and then into poker. A 14-year-old with attentive parents needed a workaround. The answer was Bitcoin: promotional tournaments that cost a few dollars, drew thousands of players, and paid the winners about a dollar. He started there and climbed. What stayed with him was the community. Strangers were willing to stake him for a full Bitcoin buy-in, never having met him, splitting profits if he won. That much mutual trust, he says, still gives him hope about people acting in one another's interest.

Fridman asks the ritual chess question. If you dropped every player in history into one tournament, Schiller says Magnus Carlsen would win, and that is not controversial. Political importance is Bobby Fischer, still the name a stranger on the street might know, an American who stood up to the Soviet machine in the Cold War and later fell apart. Among contemporaries of their own eras he offers Paul Morphy, a monster who was simply better than everyone around him. His personal favorite is Mikhail Tal, ferocious and tactical. Schiller himself was better at slow, methodical chess, but he was always attracted to creativity, and chess does not reward creativity the way building a company does. That is part of why he grew out of it. On poker he names Phil Galfond, who democratized the mathematical grind through a website, and Daniel Negreanu, who reads hands in public better than almost anyone — including in a million-dollar heads-up loss to Doug Polk that was, Schiller says, still a pleasure to watch.

Asked what a 21-year-old can tell other 21-year-olds, Schiller starts with a filter. If you feel you have to build the thing, and you can imagine doing it for ten years — because if it works, you will have to — then take that feeling seriously. Then listen to as few people as possible. Experts know their domains. They are not the authority on what is going to make sense socially to your peers. You are. Fridman is harsher. Yale teaches students, explicitly and not, to listen to the oldest person and the highest pay grade, and that is how innovation dies. A billion-dollar investor will sound right. The hard move is not "I'll hear him out and know why he's wrong." Sometimes you cannot say why. You still have to believe you are right. It looks like arrogance. It is also necessary, because many real things cannot be put into words until they exist.

Reasonable advice is the dangerous kind, Fridman says, an earworm, a radio song you hum in the car. Reason is a bad drug. Every major innovation looks impossible at birth. Love is the same mechanism: it is easy to list divorces and wasted money, and none of that helps a person who is actually in love. The whole point of an education is to walk you to a place, different for everyone, where your job is a personal revolution. Nobody will tell you that, because they want you on their path. Schiller had a private version of the scene in *Good Will Hunting*. In a graduate combinatorics class, working on discrepancy with a student who had citations, startup experience, and a later turn as CTO of a unicorn, he said he wanted to build this thing and had no technical background. The classmate, who had worked under Dan Spielman, turned and said, "I think you could do it." Schiller thought: if he says I can, who am I to say I cannot? Fridman treats it as a duty of mentorship. When a kid with a crazy idea looks up to you, say it. If you say finish your homework, you may kill the thing. Robin Williams, he adds, was a special person.

The mortality question is not hypothetical for him. Three or four years earlier, training six hours of squash a day and thinking he might walk onto Yale's team, he woke up sick, skipped practice to get a driver's license before college, and came home with a headache and spinning rooms. Red splotches appeared on his arms. The hospital sent him home with scarlet fever at 3 a.m. The next morning he felt a spear through his chest. They did not want to admit a healthy-looking 18-year-old; he insisted. A virus had attacked his heart. Myocarditis and pericarditis. At one point, doctors put his chance of dying at 25 to 35 percent. He spent about three weeks in a hospital bed, crying from pain rather than emotion, sitting forward all night because leaning back made his heart hit his ribs. For a year at Yale he could not walk to class, lived in a medical single, took shuttles, asked professors to move rooms, could not carry a bag or climb stairs, and watched muscle fall off under catabolic drugs. He started meditating, five minutes a day, up to forty, for about two years, and asked what he actually cared about.

The answer, as he tells it, was that there is an end, that some of the things he believes are not popular, and that if his body allowed it he wanted to do something that made him feel whole. Fridman thinks a struggle like that, if it leads to success, humbles and empowers at once, and tells him, not entirely joking, that he should still play squash. On the meaning of life, Schiller refuses a slogan. Part of the answer is that we get to struggle with the question together. Sometimes he thinks there is no meaning; sometimes thinking about the phrase is enough. He is young, and he hopes he will believe something different in twenty years. What does give his life meaning is conversations like this one, with people who are serious, and the chance to connect. Fridman hears an echo of the product: long-form talk is why he fell in love with podcasting, and Schiller still thinks letters can be stronger. He writes them to Russia. Put enough of yourself in the words, and you send more than semantics. You send who you are.