Curos is coming out of stealth
A group of researchers and engineers spent a year asking whether AI could be built as a public good. Today we can tell you what came of it.
For the last year, a small group of us has been working on a question that has no obvious answer: what does it look like to build serious AI as a non-profit?
We are researchers and engineers who have spent time inside commercial labs. We know how the incentives work there — the growth targets, the quarterly cadence, the product decisions made for reasons that have nothing to do with the people using the product. And we have watched as the most important technology of this century consolidated into the hands of a few companies whose duty is to their shareholders.
That is not a conspiracy. It is just how those organizations are structured. But it means the people who will live with AI’s consequences have the least say in how it behaves.
So we started Curos. Not as a side project, and not as a nonprofit arm of something else. As an actual non-profit AI research organization, funded the way non-profits are — by people who believe in the mission — with a charter that locks in what we can and cannot do.
Over the coming months we will talk about what we are building, how we are funding it, and how we are governing it. We expect to be challenged on all of it. We built Curos to be checkable, not to be believed.
This is the first time we are saying any of this publicly. We are glad to be saying it at all.
All news →