🎓 Train People on What Is Allowed

Our people break the AI policy because they have never read it. Build me a training app. Keep short modules, each covering one practical area such as what data may go into an external tool, when a human must review output, how to record a run, and what to do if something goes wrong, each built around realistic situations from our own work rather than abstract principles. Include a few check questions. Record which person completed which module and when, by role and division, and flag anyone who has not, particularly new joiners. When a policy rule changes, mark the related module as requiring a refresh and notify everyone affected rather than waiting for an annual cycle. Add an AI agent that turns a policy rule into a short training example showing a good and a bad case, so writing training is not a project, and that summarises which questions people get wrong most often so I know where the policy is unclear. Automate two things: reminders to anyone with outstanding or expired modules, and a scheduled monthly completion report. Two logins: a person who completes their own modules, and a governance owner who sees everyone.