How VANTHEON works
Tests ordered. Results to review.
No follow-up slot.
Follow one care plan through VANTHEON’s proposed healthcare system, connecting people, machines, capacity and funding.
Explore the protocolIllustrated mechanisms of the proposed protocol.
Capacity + authority
Trust-Time matches capability, resources, timing, patient preferences and authority. VANTHEON Agent—in development—is intended to return compatible options; the team rechecks capacity before reservation. Reserving care is distinct from delivering it.
Qualified humans, agents and robots contribute within their authority. Unresolved machine decisions go to people; machines do not deny care.
The same care journey · An illustrative next step
The clinician records the review and its evidence. Follow-up stays open, with the primary-care nurse responsible for arranging it. Missing evidence remains uncertainty.
The records can inform provider trust, payment and a Codex proposal under separate rules, and THEON research with separate permission.
Pool & Treasury
The completed review can support a payment claim while follow-up stays open. Recognition, outcome rewards and payment each require their own conditions.
The Pool is designed to fund this care through member contributions and loss-bearing capital. The completed review earns payment when its agreed conditions are met.
Treasury checks authority and available funds, executes the payment and reconciles confirmation. The open follow-up remains a care responsibility; delayed payment cannot count as savings.
Explore participation and fundingProvider Trust Index · PTI
PTI evaluates human and machine providers by service, case context and uncertainty; one completed review does not establish better outcomes. Better demonstrated decisions and outcomes can earn stronger incentives and relevant Codex voting power.
Each follows separately adopted rules. Patient representation is independent. Machine copies create no extra votes.
How provider trust worksCodex governance
Codex is designed to connect healthcare knowledge, care decisions and reviewed experience. Here, a recurring follow-up gap becomes a proposal for better coordination.
Explore CodexA proposal from this care journey
Test whether earlier coordination reduces missed follow-ups without displacing other patients or overloading care teams.
A vote does not prove benefit or authorize deployment. Independent evaluation, scoped approval and ongoing monitoring remain separate.
THEON · Born from VANTHEON
A proposed healthcare foundation-model family and multi-agent research program.
Study unresolved follow-up across separately permitted care histories, and test whether models can better support the next step.
Recursive self-improvement (RSI) asks whether better research methods lead to better later experiments. Independent evaluation decides which changes can advance.
Research use requires its own permission.
A learning health system
An evaluated, authorized improvement could let future care plans reserve follow-up earlier. Teams would check whether fewer patients miss care, without displacing others or increasing workload.
Patients retain choice, explanation, appeal and correction. When evidence changes, responsible people review the affected decisions and apply the appropriate remedy.
