Example Co. — Employee AI use policy
1. Purpose
Example Co. wants everyone to use AI to do better work — safely, fairly and in line with the law. This policy explains which AI tools you may use, what data may go into them, when to say you used AI and who to ask.
It applies to all employees, contractors and temporary staff from 2026-12-01. It is owned by the Chief People Officer.
2. Approved AI tools
Use only these AI tools for work:
• Enterprise AI assistant (company login)
• AI features inside our HR system
• Meeting transcription in our video-call tool
Need another tool? Ask via #ask-ai channel or ai-help@example.com. Do not use personal AI accounts for work tasks.
3. What data may go into AI tools
• Public information may be used in any approved AI tool.
• Internal information only in approved enterprise AI tools, never in personal accounts.
• Confidential business information only in approved enterprise AI tools, never in personal accounts.
• Personal and sensitive data (employees, candidates, customers) must never be entered into AI tools.
When in doubt, treat information as the stricter class.
4. Check, own and disclose
AI can be wrong with confidence. You are responsible for anything you create or decide with AI: check facts, numbers, names and tone before you use it.
Disclose AI use whenever AI-assisted content goes to customers, candidates, partners or the public, and when colleagues ask.
5. Decisions about people
AI never makes a final decision about hiring, promotion, pay, discipline or dismissal. A trained person reviews the information and decides.
AI tools that screen, rank or score candidates or employees need approval from HR and legal before use, including bias testing and notices where the law requires them.
6. Not allowed
• Using AI to infer the emotions of colleagues or candidates, or to monitor people secretly.
• Entering data into AI tools against the rules in section 3.
• Presenting AI output as checked work when you have not checked it.
• Using AI to create content that harasses, deceives or discriminates.
7. Training and support
Complete at least 5 hours of AI training for your role before using approved tools for work tasks.
Questions, ideas or concerns: #ask-ai channel or ai-help@example.com. Report any mistake involving data or AI output there straight away — early reports help everyone.
8. Employee representatives
We consult employee representatives (for example the works council) before introducing AI that changes how work is monitored, evaluated or organized.
9. Review
The Chief People Officer reviews this policy at least twice a year (next review by 2027-06-01) and publishes a short note on what changed.