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.
Two to four sentences, stated the way you would put it to the decisionmaker.
The facts that drive the issue, including the ones you wish were not there.
Who decides and whose law governs.
What you are relying on, and what the other side has cited so far.
The stage and the decisionmaker. A jury hears a counterargument differently than a judge on summary judgment.
Watch for these throughout
The model invents authority for both sides with equal fluency. Pull and Shepardize every citation before you rely on it or plan around it.
A counterargument you can knock down easily is worthless here. If the output reads comfortable, it was strawmanned. Run it again with the worst facts foregrounded.
Concede-and-pivot is sometimes the right call. Do not reject it reflexively because it feels like losing ground.
This analysis is candid about your weaknesses. Keep it in work product, mark it accordingly, and do not paste it into anything discoverable or into a client email chain that may be produced.
The killer fact often sits in a document nobody has read closely. Confirm what the record actually says before you build a response around the model's version of it.