Healthcare
Administrative and clinical settings are not the same decision
Most proposals in healthcare are administrative — scheduling, coding, correspondence, capacity — and most of the difficulty comes from proposals that quietly cross into clinical territory. The assessment draws that line early and explicitly, because the obligations on either side of it are entirely different.
Where AI is most often proposed
These are the use-case families the pack recognises. Appearing on this list is not an endorsement — several of them resolve to a no-AI outcome more often than their sponsors expect.
- Patient access, scheduling and appointment management
- Clinical coding and documentation support
- Capacity, demand and flow forecasting
- Correspondence, referral and letter processing
- Revenue cycle and claims administration
- Workforce rostering and rota planning
What actually constrains deployment here
Anything informing a clinical decision
Diagnosis, triage, treatment selection and risk stratification carry medical device obligations, clinical safety governance and professional accountability. DUN-AI assesses whether those conditions are met; it does not provide clinical judgement of any kind.
Patient data
Health data attracts elevated protection almost everywhere. Secondary use, training data and cross-border processing are examined individually rather than treated as one consent question.
Clinical safety governance
Where a clinical safety regime applies, a named clinical safety officer and a documented hazard assessment are prerequisites, not deliverables to be produced later in the programme.
Equity of outcome
Performance that varies across patient populations is a safety issue in healthcare, not only a fairness issue, and is assessed as one.
Outcomes that come up more often than expected
- Retain the human-led process, where the proposal has drifted into clinical judgement
- Approved with conditions, for administrative use cases where safety governance must be evidenced
- Data foundation first, where record quality will not support the claimed outcome
These reflect what the framework tends to conclude given the constraints above. They are not statistics drawn from completed engagements, and no such figures are published because none have been gathered.
Content status
| Property | Value |
|---|---|
| Version | 1.0.0 |
| Maturity | Foundation |
| Validation status | Unvalidated |
| Validated by | Not validated by a named specialist |
| Functions | 17 |
| Processes | 33 |
| Use Case Families | 28 |
| Questions | 20 |
| Hard Gates | 10 |
| Risk Patterns | 12 |
| Controls | 0 |
| Alternatives | 0 |
| Kpis | 25 |
| Financial Templates | 0 |
| Implementation Patterns | 0 |
| Assessment Templates | 5 |
Internally authored by DUN-AI and not externally validated. No named specialist has reviewed this content.
A clinical safety officer and a health information governance specialist should review this pack before its clinical content informs a decision.