We build simulation engines for human behavior.
Our thesis is that the biggest problems left are social. Not physical, not technical, not legal. In most cases the technology works, the capital is available, and the legal path exists. What is missing is the human system that would carry it.
A few kinds, and roughly what they cost.
Information asymmetry. Two parties hold complementary halves of something valuable and never find each other. A cancer lab has the assay a pharma program needs, and neither knows the other exists. Nobody can measure what this costs, which is the point.
Adoption lag. A medical finding takes around seventeen years to reach ordinary practice. It was proven the whole time.
Opposition. Nuclear plants, transmission lines, housing, datacenters, mines, etc. Developers spend nine figures on projects that die in hearings, and a dead project is a total loss rather than a bad return. The US interconnection queue is now measured in terawatts and years.
Coordination failure. Everyone would be better off if the group moved, and nobody wants to move first. Most of the standards, treaties, and shared infrastructure that do not exist, do not exist for this reason.
We think all of these have the same cause. People do not evaluate options and choose. They copy the people they take seriously. So an outcome depends less on whether a thing is good than on who adopts it, in what order, and who is watching.
This is not a soft claim. It has been measured for decades under names like prestige-biased transmission, complex contagion, and threshold models of collective behavior. We are building the systems that make it computable.
Artificial populations. Each person gets a history, relationships, incentives, beliefs, and a set of people they imitate. A customer introduces something new: a product, a price, a policy, a project, a candidate. Then we run it.
Say you are launching a drug. You want to know which twelve prescribers make the difference, whether the second wave arrives at month four or month nineteen, and what happens if the wrong specialty adopts first. We tell you who moves first, who makes the decision credible, where it stalls, what becomes contagious, and what would have to change to get a different result.
Every model in commercial use today does the opposite. It averages people together and fits a curve to history. The event worth predicting only happens when people stop acting independently, so averaging removes the mechanism that produces it. Worse, these models are calibrated on the stretches where nothing happened, which makes them confidently wrong at the moment they matter.
Real data gives you the one path history took. Simulation gives you the paths it could have taken.
Consumer goods and retail first, because the budget already exists. Every launch and price change is tested today with surveys where people misreport and focus groups of twelve. The shifts that matter now are social rather than individual: drinking down across a generation, GLP-1s cutting into snacking, cannabis substituting for beer. None of these can be surveyed, and none can be tested small. The research industry is over $100B and almost none of it predicts anything.
Pharma is the purest version. A drug's commercial outcome is decided by prescriber adoption, and physicians copy the physicians they respect. Launch strategy, opinion leader selection, and formulary sequencing are all diffusion problems, solved today with intuition and headcount.
Infrastructure developers need to know which coalition makes a project survivable before they spend the money. Utilities forecast rooftop solar, batteries, heat pumps, and EVs with curves that cannot represent contagion, then file those forecasts with a regulator. Insurers price behavior on tables built from a past that has stopped predicting the present. Funds take directional positions on what a population does next. Defense organizations model deterrence, which is a belief problem, with consultancies and slide decks.
The question differs. The problem is identical: act on a population once, with no way to rehearse it.
We sell predictions first.
Then we sell the population itself, rather than a report per question. It persists, it takes on the customer's own history, and every real outcome recalibrates it. A report gets consumed and the vendor gets replaced. A population compounds.
Then intervention design. Once an outcome is predictable, the next question is what would have to change to get a better one: the messenger, the order, the coalition, the price, the first group.
Then coordination. When two institutions hold complementary capabilities and cannot safely discover or trust each other, we structure the search and the exchange, and take fees, licensing, or equity in what comes out of it.
Eventually we stop selling and take positions. Equity in the adoption we accelerate, participation in what gets built, ownership of the thing that spreads. Selling knowledge earns a fee. Owning the outcome earns the outcome.
Two reasons, and neither one is difficulty.
Forecasting was built to answer what a person wants. Survey, segment, extrapolate. All of it treats a person as an independent draw, and a person is not an independent draw. He is watching someone. The unit of analysis has been wrong for fifty years.
The theory, meanwhile, has been finished and unused for decades. It was never installed inside anything predictive, because an agent used to be three rules and a probability table, and imitation between agents that simple produces nothing. That changed in the last two years. Nobody has picked it up.
The hard part is not the simulation. It is showing that a simulation predicts. We intend to be scored in public, continuously, against outcomes we did not see, until the track record is the product.
Surveys record what people say. Transactions record what they already did. History records the single path the world took.
We record populations meeting a new decision: the intervention, the prediction, the action, and the result. Plus the worlds that did not happen, where a different group moved first, or the price or the timing or the messenger changed.
This does not exist anywhere and cannot be bought, because it is not a measurement of the world. It is a measurement of a decision that had not been made yet. Foundation models will need it to reason about anything social, and any serious multi-agent simulation will need it as ground truth. We intend to be the source.
An idea nobody copies is a delusion. The same idea, copied widely enough, becomes true, and then it gets called a currency, an institution, a nation, a market. The idea did not change. The number of people taking it seriously did, and the order they did it in.
Money works this way. So do brands, bank runs, political movements, and most institutions. Something imagined changes behavior until it is materially real.
Asimov called the study of this psychohistory. He assumed one person could not be predicted and a civilization could. We are trying to build the first working version. We want to systematize, game, and control mimesis, and to introduce new ideas to the world for adoption.
Email: ryaanaqid at gmail dot com