When I was eighteen, in my first months as a journalist in Abuja, I went looking for a set of fire trucks. They had been bought by the government for more than two billion naira and then, somehow, abandoned. I found them the way you find most things no one wants found — by looking. I pulled up satellite imagery and traced the same corner of a national stadium backward through the years, and there they were: the same trucks, in the same spot, rusting in plain view, for nearly two decades. Not hidden. Not destroyed. Simply unrecorded.
That was the first time I understood something I have not been able to stop seeing since. The trucks had always been there. The knowledge of them was available to anyone patient enough to look. What was missing was not information. It was the act of drawing what was true out of the world and into a record that someone could be held to.
That same year I went looking again, this time inside the country's own laws. I had been struck by how many people my age seemed indifferent to an election that would shape their lives, and I wanted to understand why. What I found instead, written plainly in the electoral framework, was a provision for citizens who would come of voting age — and no process anywhere to actually register them. By my modeling, somewhere between one and two million eligible young people were quietly disenfranchised every cycle. The discrepancy was sitting in the founding text the entire time. Anyone could have derived it. The piece I wrote naming it was refused publication, and the gap it described remains open today.
One stadium. One generation’s voice. The same shape both times: the world held the truth, and no institution had done the work of extracting it, writing it down, and standing behind it.
I would like to tell you I learned this only as an observer. I did not. A few years later, an institution I trusted made a series of consequential decisions about my own future — decisions that turned out to be wrong — and kept no warranted record of how it had made them. Because no governed record existed, there was nothing to answer to. People moved on. A correction could be quietly filed as though it had always been there. When changes were finally made, they were made without ever acknowledging what had happened, because acknowledgment would have required a record of it, and none had been kept.
The institution was not cruel. It was something I have come to think is more common and more dangerous than cruelty. It was unknowing. Its reasoning did not outlast the moment that produced it. And so it could fail a person completely and remain, to itself, blameless — because the warrant for what it had done was never written, and what is never written cannot be inherited, challenged, or repaired.
I have spent the years since following this single thread through every domain I have touched — journalism, elections, sport, research, building software. It is always the same thread. And underneath it, eventually, I found a claim simple enough to be a foundation:
Reality always exceeds
its record.
The world is always larger than what has been written down about it. It generates more — more event, more structure, more consequence — than any record can hold, and the instant a record is made, reality has already moved past it. The distance between what is and what is kept is not a flaw to be fixed once. It is the permanent condition under which all knowing happens. Institutions live or die by how they treat that distance. The ones that do the patient work of closing it — extracting what is true, warranting it, keeping it so the next person can inherit it — accumulate knowledge and earn trust across generations. The ones that don't, fail. At the scale of a stadium. At the scale of an election. At the scale of a single life.
None of this is new. It is the oldest pattern in civilization. It is why law is written down and not merely pronounced, why a verdict comes with an opinion, why science publishes its method and not only its result, why scripture and constitutions and the common law all encode their reasoning so that those who were not present can still inherit it. Writing — durable, attributed, answerable — has always been the substrate of everything that lasts. The things we fail to write down are precisely the things that vanish when the people who knew them leave the room.
What is new is the speed and scale at which this is now happening. We have built machines that can read and write and reason over the record at a volume no institution has ever had to govern. Used well, they can finally do the patient extraction that was always too slow and too costly to do by hand — they can sit with the world's complexity and draw it into form. But pointed carelessly, they do something worse than nothing: they produce the appearance of knowledge without its warrant, conclusions stripped of the reasoning that made them trustworthy, at a scale and confidence that buries the gap instead of closing it.
Most organizations employ knowers.
Almost none are knowing.
The difference was never talent, or scale, or ambition. It is whether the reasoning behind what an institution does — not just its outputs, but the warrant beneath them — is governed, traceable, and inheritable by the next person, the next machine, the next generation that was not there when the thinking happened. A knowing institution is one whose reasoning outlasts the people who did it. That is the whole distinction, and almost everything else follows from it.
For most of history the scarce thing in any serious work was the ability to execute — to build the thing, to do the labor, to carry it out. That is changing in front of us. When machines can execute, what becomes scarce is something older and quieter: the ability to make contact with reality and say truly what is there. To make the imprecise precise. To derive the right question before the answer is generated. To stand behind what is written.
This is, I have come to believe, a golden age for a particular kind of person — the one who cannot help but pore over the world’s intricacies, who grapples with complexity under their sheets and in their sleep; the one who painstakingly makes the imprecise precise and dreams of a future worth inheriting. The writer, in the deepest sense of the word. Not because writing is suddenly important, but because everything else has been automated away from underneath it, and what remains exposed is the thing that always mattered most: the quality and the warrant of what we choose to set down.
I am building Tsari to be the institution for this. Its work — its only work — is governing the distance between reality and the record of it: drawing what is true out of the world, warranting it so it can be trusted, and keeping it so it can be inherited. Not a faster way to produce more. A more accountable way to know. The layer that makes the difference between an organization that employs knowers and one that is, itself, knowing.