The Blue-Painted Lever (*June 16, 2026*) Sometimes someone I know is deciding what to do next. Usually they are already somewhere near the work. An org, a lab, a funder, a policy place, a research group, something with the right smell. And they are tired, or bored, or unconvinced by the thing they are doing. Maybe they dislike their manager. Maybe the work feels fake. Maybe the work feels real but bad for them. Maybe they have been there long enough that “should I leave?” has become one of the tabs permanently open in their mind. So they ask around. And then the room starts producing nearby options. There is this org. There is that team. Someone is hiring. Someone is leaving, which means there may be headcount. Someone knows the founder. Someone heard the culture is much better now, or much worse now, or always depended on which desk you sat near. Someone says the pay is bad but the people are unusually good. Someone says the pay is good but you pay for it in other ways. A lot of this is useful. I don’t want to pretend otherwise. People should care where they will thrive. They should care whether a job will grind them down, or make them stupid, or make them lonely, or quietly replace all their live thoughts with institutional slogans. They should care whether they can do good work there. They should care whether they can remain a person. I’m glad people talk about fit. But I’ve started noticing a strange boundary around the conversation. Inside the boundary, everything is concrete. People name teams, buildings, salaries, managers, dinner parties, Slack cultures, grantmakers, rumours, failure modes. Outside the boundary, things become mist. “You could do something else,” someone says. And everyone nods, because this is true in the abstract, and then they go back to the real options. The strange thing is that “something else” contains most of the world. It contains normal companies, startups, consulting, academia, making money, learning a skill, moving somewhere cheaper, writing, leaving for a year, working on a problem that does not already have the right label on it. It contains many things that might be better, not just for the person, but even by the lights of the thing they say they care about. But it usually does not get the same treatment. No one says: here are three specific teams at ordinary companies where you would become much stronger. Here are five businesses you could start. Here is a boring institution with actual leverage. Here is a path that looks spiritually unflattering but gives you money, competence, and freedom. Here is the person you should talk to. Here is what it would cost. Here is why it might fail. Instead, the outside option stays outside. I don’t think this is mostly because people have compared it and rejected it. I think, often, they have not compared it. Or they have compared it in the way you compare a place you have lived to a place you have only seen from the motorway. It makes sense. The nearby options are near. They come with faces attached. They come with shared language. They come with people who understand why you are worried. If you take one, you remain in the same moral weather. You keep going to the same parties. You keep being the kind of person whose work is easy to explain to the people you most want recognition from. If you leave, maybe nothing dramatic happens. But maybe you stop being around for things. Maybe you are invited less. Maybe your thoughts become less legible. Maybe people assume you burned out, or sold out, or never really understood. Maybe you start to feel slightly ridiculous caring so much about a problem no one around you can talk about properly. Maybe you find yourself translating your inner life into a language that makes it smaller. None of this is fake. It is not shallow to want your work and your world to touch. It is not stupid to want to be around people who understand what you are afraid of. Some kinds of thinking are hard to do alone. Some kinds of seriousness need company. But I think this makes people easier to fool. Not usually by villains. Villains are overrated. Mostly by fog, status, desperation, and their own need to be holding something. There is a kind of impact case that is not exactly bad. It is just undercooked. It gestures at something large and urgent. It has the right nouns in it. It is near people who seem serious. It sounds like the sort of thing one would do if one were the sort of person who had not given up. And because of that, it gets treated generously. If the same argument were made for a normal job, people would tear it apart in twenty minutes. They would ask what the actual mechanism was. They would ask what success would look like. They would ask whether the institution had any reason to use the work well. They would ask whether the person would learn anything, or build anything, or become more capable. But put the same vagueness next to a sacred problem and it begins to look like humility. “Of course the path is unclear. The world is complicated. We’re dealing with deep uncertainty.” Maybe. Or maybe no one knows where the lever goes. I sometimes imagine these options as levers painted a pale, sky-blue colour. They sit in a room full of other people pulling similar levers. The room hums with purpose. There are diagrams on the walls. There are people you respect, or at least people you are afraid might be right. Someone tells you the lever is part of a larger system. The pipes disappear into the ceiling. You can pull it. It feels good to pull it. It feels much better than standing in the room with your hands empty, saying, “I don’t know.” The problem is not that every blue-painted lever is fake. Some of them probably do connect to something. Some may connect to exactly the machinery they are supposed to. Some may connect in ways no one can explain cleanly in advance. But some are attached to nothing. Some are attached to machines that mostly feed themselves. Some are attached to machines that convert anxious, capable people into legitimacy. Some are attached to machines that make the problem worse while giving everyone involved the warm sense of having remained loyal to the problem. You cannot tell which by looking at the paint. And this is why comparison matters. Not the fake kind of comparison where you glance briefly at the outside world, find it spiritually cold, and return to the options your friends understand. Real comparison. The annoying kind. The kind where the unbranded option gets names, numbers, paths, costs, and failure modes. The kind where you separate “this helps” from “this lets me stay close to the people who care.” The kind where you admit that fit, status, fear, and belonging are benefits, but do not sneak them into the impact column. Maybe, after doing that, you still choose the nearby thing. Fine. Sometimes the nearby thing is right. Sometimes the label is accurate. Sometimes the community is part of the work. Sometimes the vague plan is vague because the terrain is genuinely hard, not because people are hiding from the comparison. But if the case for a path gets much weaker when you remove the social warmth around it, that seems worth knowing. If the argument for the work depends on not naming alternatives, that seems worth knowing. If “impact” is doing less work than relief, belonging, and aesthetic coherence, that seems very worth knowing. I don’t want people to stop working on important problems. I want them to stop confusing proximity with contribution. I want “something else” to become real enough to compete. Because the world is large, and the room is not the world. And if you are going to spend years pulling a lever, you should at least ask where it goes.