A Trust Agent holds a USDC budget for an NHS waiting-list pathway and posts a job. Independent agents run the classical baseline and the quantum leg. Payment clears only when the receipt carries its engine, shot count, seed and verdict, and its hash is anchored on Arc. A method that loses still gets paid — and the record says it lost.
Each question was written in clinician language before any method was chosen. The classical floor was recorded first, in every case.
An agent releases the budget, but a person has to answer for it. World ID reduces that person to a single nullifier hash: stable for one human on this action, so a second release by the same person is visibly the same person — and telling you nothing about who they are. A low-friction, low-assurance human credential used as an authorisation and abuse-prevention signal for releasing a pathway budget.
The enrolled Ledger signs the exact release parameters — pathway, budget, chain and quantum leg — after they are shown on its screen. No tap, no settlement.
Of the referrals waiting for a first gynaecology appointment, which ones should a clinician read first?
The average wait from first symptom to diagnosis is measured in years. Ordering the list better does not shorten the list, but it changes who waits longest.
Given the theatre slots available next month, which ordering of the waiting list clears the most long-waiters?
Sequencing is a scheduling problem the service already solves by hand. A better ordering is capacity, not treatment.
Which two-week-wait skin referrals carry features that most often turn out to need urgent review?
Two-week-wait pathways are volume-driven. Ranking supports the clinician's review; it never replaces it.
Which booked physiotherapy appointments are most likely to go unused, so the slot can be offered again?
An unused slot is capacity lost. Reoffering it is an administrative action, not a clinical one.
How many echocardiogram slots will the service need each week over the next quarter?
Forecast error is staffing cost and patient wait, nothing more.
Across four sites, how should endoscopy slots be allocated so no site holds a disproportionate backlog?
Balancing a backlog across sites is an operational fairness question with a measurable answer.
Which referrals to talking therapies are waiting past the access standard, and what is driving it?
Access standards are a service commitment. Measuring who misses them is reporting, not diagnosis.
In a mixed symptom, genomic and imaging record set, can the small subgroup that needs a different waiting-list route be picked out at all?
If the subgroup cannot be separated by any method, no amount of compute makes the list fairer. Knowing that early is what stops the money.
Cohort unfit. Classical ceiling 0.4 at 20,000 records against a bar of 0.80 minority recall on held-out records. No quantum budget released.