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.
The legal question you need answered, in the form it was asked.
The law that governs, plus any choice-of-law wrinkle you already see.
The facts that drive the analysis. Flag anything not yet confirmed.
Cases, statutes, or secondary sources you have already pulled.
What someone does differently depending on the answer, and when.
Sets the register, the depth of citation, and how much doctrine to explain.
Watch for these throughout
The model fabricates citations, pin cites, and holdings that sound exactly right. Verify every authority in Westlaw or Lexis, including whether it is still good law. The model cannot check subsequent history.
Confidence calibration is your job. The model defaults to sounding certain, and a memo's confidence rating is the part a partner relies on most.
Doctrine shifts. Fiduciary duty, personal jurisdiction, arbitrability, and choice-of-law rules have all moved in the last few years, and the model may be describing the prior rule fluently.
A written memo is a document. Consider whether privilege attaches and survives before this is forwarded to business people or attached to a board deck.
Do not paste privileged client communications or unfiled work product unless your firm's AI policy and the engagement terms permit it.