The Manager's Guide to AI Conversations
About this resource
The most consequential AI conversations in any organization are not held in the boardroom. They happen one to one, usually unscheduled, usually when a manager is on the way somewhere else. What gets said in those ninety seconds determines whether the person tells you the truth about their AI use for the next year. This guide covers the six AI conversations managers actually have, from "am I being replaced" through to the employee who has quietly stopped trying.
Each conversation includes what the person is really asking, which is usually not the question they said out loud, then specific language that backfires and specific language that works, with the reasoning behind both. It closes with an escalation section for the questions a manager should not answer alone and a pre-conversation checklist. It is built by a licensed clinical social worker and a CPA, which is why it treats the human question and the business question as the same conversation rather than two.
The reaction in the first ninety seconds after somebody raises a hard question sets the price of the next one, for everyone watching, for months.
How to use it
Read it before you need it, then put the phrasing into your own words. Delivered verbatim these read as a script, which undoes the point of saying them. The value is in knowing which sentence you were about to say would have cost you something, and having a better one ready in the moment rather than thinking of it in the car afterward.
Inside the download
Why "nobody’s job is going anywhere, I promise" is a promise you cannot make and what to say instead.
How to separate what has been decided from what is still open, out loud.
Why "you should be excited about this" only teaches somebody to hide their concern.
The private conversation that works, and why a group setting is the worst possible venue.
Common questions
Do not promise something you have no authority to promise, because if it breaks once nothing you say afterward carries weight. Separate what has been decided from what is still open, commit to what you can actually control, such as telling them directly rather than through rumor, and then ask what specifically they are most worried about losing.
Avoid promises about job security you cannot keep, instructions about how to feel such as "you should be excited," dismissals like "it is just a tool, do not worry," and any joke about replacement. Each one either exceeds your authority or teaches the person that raising a concern with you is not useful.
Privately, and by asking rather than telling. The people who have not adopted are often the most experienced staff, who have the most competence to lose in public. Ask what would make it easier to start, and never raise it in a group setting where the question functions as an accusation with a question mark on the end.
Free. No email list, no follow-up sequence.
More from the resource library, chosen because they pair well with this one.
Nine things that decide whether any of this takes hold, and a ten-item leadership self-check scored on evidence rather than intent.
Who builds it, who breaks it, and why nothing else works without it. A nine-role matrix covering everyone from the board to the next desk.
Common beliefs about AI set beside what is actually true, including the ones that turn out to be correct.
Tell us a little about yourself and this document is yours. We don't have an email list. We collect this so we know who's finding our resources useful, and that's it.
You're downloading
this document
Your information is kept confidential and never shared with anyone.
