A letter on why medicine needs better maps.

Medicine does not need another system that guesses faster.

It needs systems that help scientists understand disease more deeply: where biology has shifted, which mechanisms are driving the state, which patient contexts matter, and what kind of intervention could plausibly move the system toward health.

Disease is rarely a single-target problem. It is a state, a network, a timing problem, a tissue problem, a patient-subtype problem, and often a translation problem. The bottleneck is not imagination. The bottleneck is knowing what should be changed, why it should be changed, and how that claim can move toward proof.

AI should expand scientific judgment, not replace it. Our work is to build an intelligence layer that turns complex biology into explicit, testable therapeutic logic. A better map. A clearer hypothesis. A stronger path from mechanism to modality to validation.

The future of discovery is not only faster generation. It is better therapeutic direction.