AI Adoption Questionnaires
Practical, evidence-first assessments you can use to align leaders, teams, and vendors — with instant scoring and copy/email outputs. Use them independently, or as a set (strategy → intake → build → govern → operate).
- Format: Instant scoring
- Output: Copy / email summary
- Use: Alignment + decision gates
Strategy & Direction
How to use Strategy & Direction
Start here to align leaders on the “why,” define success, and decide what gets built (and what doesn’t).
- 1Use this section before vendor selection or build kickoff.
- 2Score with the exec owner in the room (avoid “paper compliance”).
- 3Export the result as your decision memo: outcome, owner, timeline, and kill criteria.
AI Strategy & Adoption Readiness (Executive)
Helps leaders avoid “AI chaos” by checking strategy clarity, ownership, ship gates, and rollout discipline.
Intake & Decision Gates
AI Adoption Readiness
A baseline snapshot: evidence handling, scope, ownership, and operational safeguards.
AI Use-Case Intake & Risk Tiering
Standardises intake and decides what can ship: assigns risk tier and maps required controls before approval.
Build & Reliability
AI Engineering Maturity (Systems)
Assesses production discipline: testing, stage-separated eval, monitoring, logs, and failure-path handling.
AI RAG & Grounding Readiness
Assesses whether your RAG system can ground claims with permitted sources, correct citations, and refusals when needed.
Governance & Procurement
AI Governance & Compliance Readiness
Checks whether enforceable controls, evidence trails, accountability, and incident readiness are in place.
AI Vendor Procurement & Due Diligence
Prevents demo-driven buying by checking logs/exportability, failure-path behaviour, data handling, and audit support.
Operate & Review
AI Change, Drift & Incident Review (Operational)
An operational risk review for changes/releases and incidents: drift signals, rollback gates, and required actions.
How to use Operate & Review
This is your “production utility”: use it when drift, change, or incidents introduce real-world risk.
- 1Run it for any release that changes model, data, prompts, tools, or policies.
- 2Run it again when drift/quality drops, or when you see unsafe behaviour.
- 3Use the output as the incident record: what happened, evidence, actions, and ship gate.
