Is Microsoft Copilot safe to roll out to our staff?
It is safe when your permissions are clean, and risky when they are not. Copilot respects existing access exactly, so it surfaces anything a user could already technically reach, including the payroll folder nobody realised was shared. REDD fixes the permission layer before deployment, which is the whole difference between a productivity gain and a disclosure incident.
What is shadow AI and why does it matter?
Shadow AI is staff using unapproved tools, usually free public chatbots, to get their work done. It matters because company data pasted into those tools may be retained and used for training, and because you cannot govern what you cannot see. The fix is not a ban that everyone ignores, it is providing approved tools that are genuinely better than the free ones.
Do we need an AI policy?
Yes, and increasingly you will be asked to produce it. Insurers, auditors and government buyers now ask how AI use is governed. A workable policy names approved tools, sets data boundaries, and defines what must never be pasted into a model. REDD writes one your auditor will accept and your staff will actually follow.
How long does an AI readiness assessment take?
The assessment itself is typically two to three weeks depending on tenant size. You get a report covering access and permission exposure, data classification gaps, and a prioritised remediation plan with indicative effort. Many businesses stop there for a quarter and fix the foundations before rolling anything out.
Will AI replace our staff?
Not in the deployments we run. What it reliably removes is the repetitive work around the edges: summarising, drafting, formatting, searching. The productivity gain shows up when people stop hunting for information and start using it, which is a training and workflow problem more than a technology one.
What does AI enablement actually cost?
There are two components: the readiness and governance work, which is a defined project, and the licensing, which is per user. Most clients find licensing is the smaller surprise. The bigger variable is how much permission remediation the readiness assessment uncovers, which is exactly why we assess before quoting the rollout.
