This page lists three turns, the messages you send to the AI one after another in the same chat, and then a check of the answer. Keep it open next to your chat. Each turn shows the next thing to paste, with the values you entered already filled in. Nothing is sent from this page, and your conversation stays in the AI tool you are using.
What is being investigated, the working theory, and where the investigation currently stands.
Role, tenure, proximity to the events, employment status, and anything sensitive.
What you hold that her account will be tested against, with dates and the specific lines that matter.
Changes the warning, the note-taking plan, and the privilege posture.
The state or country, plus the recording-consent rule if you know it.
Watch for these throughout
A muddled Upjohn warning creates real problems. Courts have found waiver and disqualified counsel over warnings the witness did not understand. Deliver it verbatim and note in the memo that you did.
Recording without the consent the jurisdiction requires can be a crime, not just a problem. Confirm the rule where every participant sits, including remote attendees in other states.
If the witness has personal counsel, coordinate scope and logistics before the interview. Contacting a represented former employee directly can create an ethics problem for you.
The model will build questions around documents you described but did not paste, and it will get the quoted language slightly wrong. Verify every exhibit reference against the actual document before you use it.
Interview memos that read as pure fact recitations are more vulnerable to disclosure than memos framed as counsel's assessment for the purpose of legal advice. Frame the memo before you write it, not after.