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.
Log notes, correspondence in and out, reserve changes, internal referrals, with dates.
The elements you want applied, with the case or statute you took them from. Do not leave this to the model.
Whose law governs the claim, and where a bad-faith suit would be filed.
Your side, which decides whose position gets stress-tested hardest.
Every demand, its terms, its deadline, and the carrier's response inside that window.
Policy limits, the demand, reserves, and a realistic verdict range.
Watch for these throughout
Bad-faith law varies more by state than almost anything else in insurance: elements, whether a statutory claim carries a private right of action, the genuine-dispute defense, punitive thresholds, and damages all differ. Supply the standard yourself and verify it; do not let the model produce one.
Whether reserves, claims manuals, internal guidelines, and coverage-counsel opinions are discoverable turns on state law and on the posture of the case. Confirm before you concede or withhold anything.
A wrong coverage decision is not bad faith by itself in most states. Analysis that conflates the two will overstate exposure and mislead a reserve discussion.
This analysis is protected work product only if you create and route it that way. Label it, and do not paste an unproduced claim file into any tool your firm's AI policy does not cover.
Time-limited policy-limits demands generate the largest excess exposure and often are not in the file you were handed. Confirm the demand history independently.