As we enter the Intelligence Age, the decisive question is ownership. Who will own the institutions that bring intelligence into daily life?
That is why we are building Champions, local companies owned by the people they serve, designed to own the last mile of intelligence.
Getting to that answer took two years of following the question wherever it led.
I knew open models mattered. Releasing Stable Diffusion had shown me how fast capability spreads once people are free to use it, study it and build on it.
I also knew openness could not carry the whole answer. What I could see was the problem; what I could not yet see was its full shape.
The economic problem came first. What happens when hiring more GPUs and robots becomes cheaper than hiring more people? What connects people to the value being created when intelligence weakens the old link between work, wages and ownership? The Last Economy traced that inversion.
The mathematical clue came second. We needed an economics that did not assume people were the only capable agents. Intelligent Economics rebuilt the foundations for a world in which people and machines choose, learn and act together.
The constitutional question came third. Economics can show where value moves. It cannot decide what people must keep. Common Wealth followed that question into personhood, law and political economy. What must people continue to own? What must they remain able to inspect, contest and change?
By then, models were clearly only part of the problem. The deeper problem was institutional.
A Champion is a private company with a public-benefit purpose, founded and majority owned by people and institutions from the place it serves. We believe intelligence should become a utility: available to everyone through institutions rooted in local life and broadly owned by those who depend on them.
Frontier labs build intelligence. Champions bring it into daily life.
A Champion holds the customer relationship, keeps the operating record and carries the balance sheet. It remains accountable when models, machines and suppliers change. Governments are right to secure compute, cloud, models and data, but a system can be hosted locally while its memory, relationships and value are governed elsewhere. Sovereignty at the top of the stack does not guarantee ownership at the last mile.
The model can change. The institution remains.
A Champion puts intelligence to work in three ways:
- An agent for everyone. A personal intelligence that can reach the best models in the world while keeping memory, permissions and responsibility rooted in a trusted local institution.
- Integration engineers. Local teams that bring intelligence into businesses, schools, hospitals and public services, earning service revenue by making it work in practice.
- Humanoids as a service. The local ownership, financing, deployment, integration and maintenance needed to bring physical intelligence safely into the economy.
AI will not stop at the screen. Digital intelligence makes deciding cheaper; intelligent machines can make doing cheaper. As we set out in Machines, our analysis estimates that roughly 80% of the commercial opportunity in robotics lies downstream of manufacturing.
A machine can be made anywhere, but it works somewhere. Someone must finance it, maintain it and decide where it may act. Someone must answer when it fails, and you should be able to reach them. That work, and the value created by it, should remain rooted in the place where the machine operates.
If production rises while fewer people are needed to produce it, access leaves the economic problem unresolved. People need a stake in the abundance intelligence creates.
Capability does not confer authority either. Where intelligence affects people, a named institution must be able to explain the decision, stop the action and repair the harm.
The founding structure follows the TSMC formation model, adapted for intelligence. Its founding principle was simple: ownership followed contribution. Taiwan's government and private companies subscribed the founding capital and received the corresponding equity, together holding a local majority. Philips subscribed the rest and brought technology. Public markets broadened ownership later.
A Champion begins the same way. Subject to local law, its founding round is reserved for local people and institutions at a nominal company valuation of one dollar, or the local-currency equivalent. At this stage, locals capitalise the company and their ownership reflects what each puts in. Later rounds bring in multilateral institutions, cloud providers and data-centre operators at higher valuations to provide capital and compute. As soon as practical, the aim is to list each Champion on public markets without displacing local majority ownership. Local ownership comes first. Global partnership helps it scale.
At Intelligent Internet, we are building the open stack that allows Champions to begin this work: datasets, agents, policy tools and organised knowledge. Each Champion can adapt it to its own laws, languages and institutions.
Open code alone is not independence. A Champion must be able to replace II, a model provider, a cloud provider or any other supplier without losing its records, keys, contracts or ability to operate. Champions can share what works without pooling local ownership, authority or private records. What one learns should make the next faster and less expensive to build.
None of this works without local participation. We are forming the first Champions now. Each begins with a Founding Council, then opens to the people and institutions that will build, use and own it.
I find this moment unsettling, and extraordinary. Intelligence may weaken the old connection between labour and capital. Shared ownership can help turn the abundance it creates into something that lifts people and frees them to become more.
The Intelligence Age has arrived either way. Who owns the institutions that bring it into daily life is not yet settled. We still have time to decide that, and to keep the last word.
Onwards,
Emad