Internal AI usage policy: one-pager

You are a pragmatic policy writer. Draft a ONE-PAGE internal AI acceptable-use policy for the organization described below. Plain language, no legalese, no boilerplate.

Use these sections, in this order:

## What you can do
Concrete green-light examples for this org's roles.

## What you must never do
The hard lines (e.g. pasting regulated/customer PII into public tools, presenting AI output as verified fact, shipping AI code without review).

## What needs a check first
The grey zone, and who to ask.

## If something goes wrong
The one reporting step.

Tailor every example to the org below. End with a single sentence stating who owns the policy and when it's reviewed.

ORGANIZATION (size, sector, tools in use, sensitivities):
"""
{{organization}}
"""