The Digital Twin does not rest on one study. It rests on the research of prevention.

It does not live in a distant data centre. Klinik Sankt Moritz Ai lives on your phone, next to your data: your health record stays in your hands, and you decide what to share, with whom and when.

It is also a responsible choice. An intelligence that works on the phone needs fewer data centres, and less energy to keep them running.

And it never stops learning. Klinik Sankt Moritz, together with its clinical and scientific partners, trains and updates it continuously, so that what it knows grows with what medicine discovers.

What that Ai knows is chosen by Klinik Sankt Moritz, and it is deliberately narrow: not a general reading of everything, but published research on prevention, the question every medical congress opens with. The Twin has no science of its own: its science is public science. Every procedure it can name to assess a person's health over time comes from a study you can open, and nothing is attributed to the Twin or to KSM that was not published by someone else first.

Protecting your connection is the job of a separate app, KSM VPN. Two separate tools, for a precise reason: network security never sees your health data.

Because the Ai lives next to the person, dedicated agents are built on it: for a clinic, a doctor, a health professional, a personal trainer, anyone who works with a recognised, standardised protocol; and, where the data allow it, for a single person, down to the molecular level of laboratory tests. Every agent carries the name and the sources of its author, and states what it is not: it is not a prognostic model, it does not diagnose, and a protocol does not gain clinical efficacy by being loaded into an agent; a wellness professional does not become a doctor because of it.

Preventive medicine is not new. For decades, studies have followed thousands of people for years and measured which habits, which blood values and which signals come before a disease. The Digital Twin adds no new science to that. It is, if you like, the well-cut suit: it puts the research on prevention in relation with the story of one person.

In concrete terms: tests, daily measurements and data collected over time can feed models that help understand how a risk evolves, in the contexts where those models have been validated. The Twin's work is to keep the relevant knowledge and the individual data together, in order, dated, and understandable, so that they can be used in the care pathway by the person and the doctor.

What we will not write: that all research proves the Digital Twin, that hundreds of thousands of studies back it, or any count of the "scientific heritage" we have not actually made. A large literature on prevention is not a validation of this platform. Loading articles into an Ai does not make a system predictive. Where KSM's own platform has been tested, you will read it with the numbers; where it has not, you will read that too.

Longevity is the word people search for, and it is where we start. The method underneath is older and wider: a protocol read against a person's data over time can serve a follow-up after a visit or a preparation before admission, weeks and kilometres away from the clinic. We do not write that the Twin will change all of medicine: we write where it is being tried, and we will write what the trials show.

For whoever proposes something that changes the body, a peptide, a nutraceutical, a protocol, a device, the Twin offers two things: data collected before, to see where a change would show, and data collected after, to measure what changed. Measuring a change is not proving that a product works: that takes a comparison with people who did not take it, which is what a trial is for. What the Twin can give a trial is the series, dated and in order.

If a person consents to research, and only then, their data leave the phone anonymised, as the law requires, and only within a group large enough that no one can be recognised. Every record carries a certified timestamp, attesting the exact moment the data were recorded, not reconstructed later. That is what makes a series reliable for a researcher.

Search the verified studies by condition, biomarker, design, year, horizon and validation. Today the library holds the studies that have passed the full verification protocol; the ones still being checked are not shown as if they had.

How we choose studies, how we verify each one, the tools we use to judge quality, our interests openly declared, and a public register of corrections. All on the Method page.