AI policy checklists

A short pre-approval list for a new AI tool or feature.

Sort information into safe, internal, confidential and restricted AI-use categories.

Check whether your policy has an owner, tool list, data rules and review path.

Questions to ask about terms, data handling, access and internal ownership.

Review generated text, recommendations and assets before they affect a real person.

Decide when your client should know that AI assisted a deliverable or workflow.

Launch, communicate, train and collect acknowledgement without overcomplicating it.

Make sure the policy acknowledgement is specific, accessible and recorded.

Quickly identify data, impact, review and escalation risks in a proposed use case.

Check claims, sources, confidentiality, rights and brand fit before publication.

Respond to accidental sharing or unapproved AI use with an accountable record.

Refresh your policy when tools, terms, team roles or data handling change.