AI Use Assessment Template

Blank template

1. Intended use and AI function
2. Decision authority and prohibited uses
3. Data, population, and reference dataset
4. Failure scenarios and critical misses
5. Human review design
6. Performance measures and acceptance criteria
7. Source verification and data integrity
8. Access, audit trail, and supplier responsibilities
9. Testing and release conditions
10. Monitoring, stop, and revalidation criteria

Completed fictional example

Fictional teaching example: This is not an executed or approved validation record.

Intended use and AI function

Draft nonauthoritative validation-protocol review comments from permitted documents

Decision authority and prohibited uses

Qualified reviewer disposes findings; AI cannot approve, reject, sign, or release a document

Data, population, and reference dataset

Approved English-language protocol types with independently established reference findings

Failure scenarios and critical misses

Missed approval, false traceability gap, wrong source version, unsupported compliance claim

Human review design

Reviewer sees source passage, requirement, and rationale and can accept, reject, or request clarification

Performance measures and acceptance criteria

Critical-finding recall, false findings, citation correctness, performance by issue type, and reviewer detection

Source verification and data integrity

Immutable source version and page/section reference retained with the potential finding

Access, audit trail, and supplier responsibilities

Role-restricted use; input, version, output, disposition, and change recorded

Testing and release conditions

Representative challenge set and human-workflow study; scanned handwritten records excluded

Monitoring, stop, and revalidation criteria

Stop on critical-miss threshold, source failure, unauthorized action, or material model/prompt/parser change