Main research programs

Open-Ended Systems. Can we build systems whose capacity for creating knowledge is genuinely open-ended, systems that keep generating new understanding rather than running out of things to learn? Most engineered systems hit a ceiling once they master a defined task. A handful do not. Science keeps producing new theories, markets keep producing new categories, language keeps inventing new ways to mean things. The frontier never closes. We want to understand what makes a system behave that way, and whether AI systems can be engineered for it. Our inaugural post, Engineering Evolution, lays out the North Star of this research.

Designing Durable Trust. Can we build the trust systems, verification mechanisms, and institutional scaffolding that let knowledge survive and compound across time? Knowledge creation without durable infrastructure is a fire without a hearth, brilliant and temporary.

The two programs are not independent. Open-ended knowledge creation requires institutional conditions that are historically rare and fragile, and durable institutions are only worth building if the knowledge they protect can keep growing.