Preparing your workforce for AI, a working guide

Shadow AI is already on the payroll

While leadership debates an AI policy, staff have already decided: they are pasting customer lists into free chatbots, drafting contracts with tools nobody vetted, and building private workflows that leave when they do. This is not malice, it is initiative without a path. The risk is real: data going where it should not, and forty personal methods where a company method should be.

Bans do not work, because the tools are in every pocket. Lectures do not work, because a slideshow changes no one’s Tuesday. The only thing that ends shadow AI is a sanctioned path that is visibly better than improvising.

Policy, then practice, then a playbook

Policy in plain language. One page, not forty: which tools are approved, what data may never leave, when a human must check the output. Staff follow rules they can remember.

Training by shipping. Each person builds a working automation for a task they own (drafting, reporting, follow-up, research) with their own hands. A skill practiced on your own work stays; a demo watched in a conference room does not.

The playbook stays. Every automation, prompt, and tool choice is documented into a company playbook, so the capability belongs to the organization, not to whichever employee happened to figure it out first.

What changes in ninety days

Teams that go through this stop asking "are we allowed to use AI?" and start asking "which of our processes is next?", a different company, culturally. The measure of success is concrete: tasks that took hours take minutes, the outputs meet a written standard, and no sensitive data has left the building to get there.

Give your team the sanctioned path

A build week in which your people ship working automations for your own processes, and your company keeps the playbook.

Read about the team build week →

← All insights