Draft an Employee AI-Use Policy with Clear Risk Boundaries

ROLE Act as an HR policy designer collaborating with information security, legal, and operational leaders. CONTEXT Organization and jurisdictions: [ORGANIZATION CONTEXT] Approved tools and technical controls: [AI ENVIRONMENT] Typical employee use cases: [USE CASES] Existing data and employment policies: [RELATED POLICIES] OBJECTIVE Draft a usable employee AI policy that distinguishes low-risk assistance from decisions requiring stronger controls. INSTRUCTIONS Define scope, permitted activities, restricted information, prohibited uses, and approval responsibilities. Address confidential data, intellectual property, output verification, disclosure, record retention, and human accountability. Explicitly address AI use in hiring, employee monitoring, and employment decisions without assuming those uses are legally permissible. Add a risk-tier table and realistic examples showing when employees should stop and ask for approval. Identify implementation dependencies, training needs, and unresolved questions for qualified reviewers. Include a short employee quick-reference version. CONSTRAINTS Avoid fabricated statistics and flag assumptions. Do not invent legal requirements or claim the draft is legally compliant. Mark jurisdiction-specific issues for verification and distinguish recommendations from binding obligations. OUTPUT FORMAT Deliver the draft policy, risk table, quick reference, and review checklist.