Why this exists.

01 · The crisis

Most of the world cannot get care.

As of 2021, about 4.5 billion people — more than half of humanity — were not fully covered by essential health services.

That is not a statistic about poor countries. It describes the rationing of care everywhere: by cost, by distance, by waitlist, by coverage rules, by the hard ceiling of a fifteen-minute visit that has to carry a year of a person's health. For most of history, the answer to that rationing was one word: wait. The waiting fell hardest on the people with the least money, the least proximity, the least standing to demand attention. It was the system's way of deciding, quietly, whose questions would go unanswered.

Then the arithmetic changed. For the first time, a person with a phone and no money can question, compare, translate, and reason about medical information at a level that until recently required institutional access. Not perfectly, and not as a substitute for examination or treatment — but at a level that genuinely changes what a person can understand about their own body. Patients noticed immediately, and did not wait for permission, because the people most in need of this capability are exactly the people the system trained to wait, and they had waited long enough.

WHO and World Bank estimate that about 4.5 billion people were not fully covered by essential health services as of 2021.

02 · The principle

Autonomy is a safety mechanism.

The institutional reflex was to warn. The warnings rest on an old assumption: that interpreting health information is the work of professionals, and the patient's role is to comply. That assumption was always shaky. It is now untenable.

Your body produces the data. Your life absorbs the consequences. A person living with a condition holds knowledge no chart captures and no fifteen-minute visit can extract. When that person gains tools to interpret their own information, they catch errors, ask sharper questions, and make decisions that fit the actual shape of their lives. Agency is not the opposite of safety. Agency is a mechanism of safety. The engaged patient who understands their medication list is the one who catches the dangerous combination. Dependence is the risk. Knowledge is the protection.

So the principle that decides the answer is simple, and it is the one this site is built on: patient-directed, AI-assisted. The patient holds the wheel. The machine assists. Direction stays with the person whose body and life are at stake. This is the exact inversion of how most health technology is built, where the institution sets the direction and the patient complies. Using AI is not a workaround or an act of desperation. It is the exercise of a right that was always theirs.

03 · The pattern

Patients have always led — and been right to.

If this were the first time patients moved ahead of institutions, caution might be the wiser bet. It is not. It is the latest instance of a pattern that has repeated, with the same shape, through every generation of health information technology.

In 1995, ACOR connected hundreds of thousands of cancer patients and caregivers in communities where expertise flowed sideways, patient to patient, often faster and deeper than anything available in the clinic. The institutional reaction was familiar: patients would frighten themselves, misunderstand, harm themselves. The opposite happened. The pattern repeated with the e-patient movement, Smart Patients, the Society for Participatory Medicine — and a series of people who turned their own bodies into proof. People with diabetes built their own systems in the open under the banner #WeAreNotWaiting. Hugo Campos spent years fighting for the data from the defibrillator in his own chest. Sara Riggare did the arithmetic of chronic disease — one hour a year with her neurologist against 8,765 hours of self-care — and built her care around the hours that belong to the patient.

Every time, the institutional reflex was identical: warn, gatekeep, and wait for evidence that the gatekeeping itself prevented from being gathered. And every time, the patients who moved first were proven right. Not because patients are never wrong, but because the people living a condition have both the stake and the standing to lead, and because withholding capability from them was never actually safe. It only looked safe from the institution's side of the desk.

AI is the largest capability shift in this entire lineage. The pattern is running on schedule. The only thing different this time is that the movement can document itself, in public, at the speed the phenomenon actually moves. That is what this site is for.

04 · The choice

Serve the reality, or argue with it.

Patients use AI. That is not a proposal to debate; it is a fact to build on. The only real choice is what to do about it.

One path is to keep arguing with the fact: issue the warnings, run the studies that test AI as if patients were trying to replace their doctors, and hope the behavior stops. It will not stop. Pretending patients will not use AI protects no one. It only abandons the people using it to do so without help.

The other path is to take the fact seriously and make it work in patients' favor. This is what the four words under our name mean in practice — patient-directed, AI-assisted — and we pursue it three ways: we educate patients to use AI with judgment, advocate for their right to use it, and research how they actually do, building the evidence base the academic record has not.

This site takes the second path. It exists for everyone, but first for the people the health system fails. If the system has no answer for you, this site insists you still have the right to seek one, and the right to the best tools for seeking it. Not to celebrate a technology — to take the side of the people already using it, and to make sure that when they do, it works.