Who it is for
For anyone who wants to use AI tools without handing over their privacy or their company's secrets.
Use AI tools without handing over your privacy or your company's secrets.

Most AI safety advice points at the wrong end of the conversation. It warns you about hallucinations, bias, and made-up citations — all problems with what the model says back to you. Those are real, but the bigger risk is what you type in.
You've built the parts: a terms read, a tier system, a redaction habit, a settings audit, a team policy, a defensibility file. But most AI exposure doesn't happen in the moments you plan for — it happens in the ones you don't.
For anyone who wants to use AI tools without handing over their privacy or their company's secrets.
You judge the risk by what you're about to share — a draft email, a spreadsheet row, a contract clause, a customer complaint, a block of code — not by which tool you're using.
You get a repeatable way to abstract a situation and strip the identifying details, so you're not doing it by hand every time — rushing, missing the client name in paragraph three, pasting the spreadsheet "just this once.".
You see why individual discipline doesn't scale: the moment a colleague pastes a customer list into a free chatbot, your careful work stops mattering. Shared data means shared risk, and rules only hold when someone writes them down and makes them easy to follow.
You're ready when a client's general counsel asks, in writing, exactly which internal documents your team entered into AI tools, under what terms, and whether their confidential materials were among them — or when a regulator asks how you handle personal data.
Open your Data Rules File and build the table. Be honest — it's a private baseline, not a performance review. List every AI tool you've used for work in the last month and what you actually fed it.
Score yourself 0–3 on each line (0 = never, 3 = always), total it, and date it. You'll re-score the identical list later to measure how far you've moved — starting with whether you know if each tool trains on your inputs, and whether you check the account type before pasting work data.
Tape this to the front of your Data Rules File. By an earlier chapter it becomes automatic; for now, run it manually.
You're a sales manager who wants AI help rewriting an underperforming renewal proposal. Watch how the input — not the request — decides the risk: the higher-risk version pastes the full renewal email.
Pick one AI tool you actually used this week. Open its terms of service and privacy policy in another window and fill in the worksheet, a sentence each: tool and plan, date reviewed, and first, whether your inputs are used to train or improve the model.

EPUB, PDF, and HTML are included so the book can work on an e-reader, as a designed copy, or as a searchable desk reference.
For e-readers and reading apps.
The designed edition with diagrams and layouts intact.
Searchable, copy-pasteable, and practical as a reference.
Yes. You get the complete edition, including the chapter sequence and internal materials described on this page.
EPUB, PDF, and HTML are included so you can read on an e-reader, keep a designed copy, or use the searchable browser version.
Because this is an instant digital download, broad change-of-mind refunds are not offered after the files have been accessed. Refund requests are reviewed within 7 days for duplicate purchases, accidental purchases before access, access failures we cannot fix, wrong files, corrupted files, or pages that materially misdescribe the book.