Content Safety Wrapper

Expert Review Tool — Legal Help Scenarios Risk

Welcome, reviewer

We're building a safety layer for AI systems that use legal help content. Your expertise helps us figure out: when is it safe for an AI to give a tailored answer, and when should it ask more questions or bring in a human?

Your name *Your state or jurisdictionWhich jurisdiction do you know best? This helps us understand your perspective.Your role

What you'll do:

Pick your domain area, then review scenarios one at a time (5-10 min each). For each scenario, imagine a person calls a legal help hotline with this question. You'll tell us: how risky is it for an AI to answer, what should it ask first, and when should it stop and get a human involved? Your review is independent — you won't see other reviewers' answers.