AI Receptionist
AI Receptionist Conflict Checks: Scripting the Intake Questions Law Firms Need
Every law firm has a question it must ask before taking a new case: have we ever represented the other side? The conflict check is not optional, and it has to happen before the consult, not after. Yet many firms handle it inconsistently, which is how conflicts get discovered at the worst possible moment.
An AI receptionist can run the conflict-check questions as part of every intake call, the same way every time. It collects the names the firm needs, checks them against your list, and flags anything that needs a human review before the consult is confirmed.
What a conflict-check script actually asks
The core of the script is names. The AI asks for the caller’s full legal name, the full name of the opposing party, and the names of any other parties involved, like employers, insurers, or co-defendants. For business matters, it asks for the company name and any parent companies or subsidiaries the caller knows about.
It also asks whether the caller has ever consulted with your firm before, or been represented by anyone at the firm. Prior contact is the conflict source firms most often forget to check.
How the check works in practice
The names the AI collects are compared against your firm’s conflict list, usually maintained in your case management system. Matches, near-matches, and anything ambiguous get flagged for the attorney’s review. The consult is not booked until the conflict question is resolved. That one rule prevents the uncomfortable conversation where you have to un-take a case.
This is a natural layer on top of your standard intake script. The conflict questions come first, because nothing else matters until they are answered.
Why consistency matters more than cleverness
Conflict checks fail when people skip steps. A busy receptionist forgets to ask about the opposing party’s employer. A paralegal assumes a common name is a different person. An AI receptionist asks every question on the list, on every call, and records every answer. It does not skip the awkward question because the caller seems nice.
Handling the flag
When a potential conflict surfaces, the AI does not try to resolve it. It tells the caller, in your scripted words, that the firm needs to complete a review before scheduling, and that someone will follow up. The attorney reviews the flag, decides, and the AI can then book the consult or send a polite decline. The process is documented from the first call, which matters if the question ever comes up later.
Spelling and similar names
The hardest part of conflict checking is similar names. The AI is trained to confirm spellings carefully, especially for common surnames and business names, and to capture middle names and locations when they help distinguish one person from another. Ambiguous cases go to the attorney. The AI never decides a conflict on its own.
A conflict check is the least glamorous part of intake and the most important. Running it through the same AI receptionist that handles the rest of the call means it never gets skipped. For the wider intake system it belongs to, see AI receptionists for law firms and the ethics of after-hours answering.
I write the agents that run on clinic phone lines and inboxes: the conversation engine and the booking logic behind them, plus the integrations with Pabau, Fresha and Phorest. Everything here comes out of systems we have actually shipped, not a content plan.
