Who this is for
- Practice managers
- DPOs
- Clinic owners
- Boards and partners
- Clinical governance leads
Common shadow AI examples
Checklist
- ChatGPT used to draft patient letters
- Microsoft Copilot used across documents or emails
- AI meeting transcription used for clinical discussions
- Personal-device AI assistants
- Free-tier transcription tools
- AI summarisation of referrals or complaints
- Marketing AI tools using patient-related context
- Supplier platforms with AI features switched on
Why it becomes an evidence problem
The risk is not that staff are interested in AI. The risk is having no documented governance position when a DPO, board, patient, insurer, MDO or inspector asks how AI use is controlled.
Evidence to create
Checklist
- Confidential, role-level and non-disciplinary AI use survey
- AI tool and use-case inventory
- Approved, conditional and prohibited use matrix
- Staff guidance on patient data
- Vendor evidence tracker
- DPIA screening records
- Incident reporting route for AI-related issues
- Board-level summary of AI exposure
What ELSA AI can help produce
The Diagnostic surfaces shadow AI use and structures the first evidence response.
Outputs include
- AI Tool and Use Case Inventory
- RAG Exposure Map
- Staff-use evidence indicators
- DPIA Readiness and Patient Data Exposure Note
- 30-Day Priority Action Plan
The issue is not simply that AI tools exist. The issue is having no documented governance position when someone asks how that AI use is controlled. Clinical AI Exposure Diagnostic™ page explains scope, timeline and fees.
Advisory governance support only. Not legal advice, DPIA sign-off, CQC certification, ICO approval, insurer coverage advice, MDO indemnity advice or clinical safety case sign-off. Final decisions remain with the clinic’s accountable officers and advisers.
Need this evidence mapped for your clinic?
The Clinical AI Exposure Diagnostic™ gives clinic leadership a board-ready view of AI use, patient-data exposure, evidence gaps and priority actions in four working days from completed intake.