How do I write an AI policy for my company?
Chris Duffy
Certified Chief AI Officer • • 2 Min Read
Keep it to one or two pages and cover four things: which AI tools are approved, what data can and can't go into them, who signs off new tools, and what happens when something goes wrong. That's a working AI policy. The forty-page version nobody reads protects you less than the short one everybody does.
Start from how your team already works rather than from a template. Watch what tools they're using now (they're using something), draw the data line clearly, and name one person who approves new use. Write it in plain English so people follow it without a lawyer.
Then keep it alive. A policy that's dated and reviewed carries weight in a client's due diligence. One written once and forgotten doesn't. Our free AI Manifesto template is the one-page version: vision, permitted and prohibited uses, data boundaries, human oversight, accountability.
Read next: AI governance for UK SMEs · What is a DPIA · Read more on our Substack · Book a call