Printable readiness checklist

Private AI readiness checklist for business teams.

Use this checklist before a pilot. Assign an owner and record evidence for each decision; a product demo alone does not prove that identity, data routing, recovery, and support will work in your environment.

Discuss a scoped pilot

Data and use case

Name the workflow, data owner, data classes, prohibited content, retention need, and acceptable model destinations. Prepare representative test material that does not expose production secrets during evaluation.

Identity and endpoints

Choose SSO and provisioning ownership, define administrator and break-glass access, inventory managed devices, decide how clients receive server configuration, and document the employee revocation test.

Models and knowledge

Approve model sources and licenses, measure memory and latency on target hardware, map every document and embedding destination, preserve source permissions, and test update and deletion.

Operations and recovery

Assign monitoring, audit export, certificate renewal, backup, restore, upgrade, rollback, capacity, and incident owners. Run a restore test and collect a redacted support bundle before production.

Pilot exit criteria

Define quality, latency, error, privacy, revocation, and recovery thresholds. Record limitations and the decision maker who can accept them. Expand seats only when the process is repeatable without engineering intervention.