Every answer AskSensai gives carries a safety flag: green, yellow, or red. Green means general legal information is sufficient for the question you asked. Yellow means your situation is specific enough that you should consider attorney review. Red means the question is safety-critical, and the answer is gated behind a human-attorney sign-off path rather than delivered as-is. The flag is not decoration. It is the platform's honest assessment of whether an AI answer alone is a responsible place for you to stop.
This post explains why the flags exist, how the system decides which one your question gets, and what each color should change about your next step.
Why does an AI legal product need safety flags?
Because legal questions are not equally dangerous to get wrong. "What is the difference between a will and a trust" is a question where a clear, sourced explanation genuinely solves the problem. "My custody hearing is Friday and my ex just left the state with our daughter" is not. The same answer engine can be exactly right for the first person and irresponsibly incomplete for the second, not because the information would be wrong, but because the second person needs a lawyer, a deadline calculation, and possibly a courtroom, not a paragraph.
Most legal information products handle this with a blanket disclaimer at the bottom of every page, which users learn to scroll past. LawSensai's approach is to make the risk assessment itself part of the answer. Instead of treating every question as equally safe, the platform classifies each one and tells you, in one color, how much weight an AI answer can safely carry in your specific situation.
How does LawSensai decide which flag your question gets?
When you ask a question in plain English, the answer engine reads three things out of your description before it composes anything:
The practice area. Whether your problem sounds like landlord-tenant, family law, criminal defense, consumer debt, and so on. This matters because risk lives differently in different areas. A pricing question about trademark filing fees carries different stakes than anything touching criminal exposure.
The jurisdiction. The state or city you mention, and if you did not mention one, it asks. Legal answers change at state lines, and so does urgency. A deadline that is comfortable in one state can be nearly expired in another.
The posture. Where you are procedurally. Someone who is curious about how evictions work is in a very different position from someone holding a summons with a court date. The same topic can be green for the first person and yellow or red for the second, because posture is what turns information into a decision with a clock on it.
Those three signals, together with what the question itself involves, produce the flag. The flag then travels with the answer, so you never get the information without the risk assessment attached.
What do the green, yellow, and red flags mean?
Green means the question is informational and the answer can responsibly stand on its own. Definitions, how processes generally work, what documents typically say, what rights generally exist. You still get the reasoning and the jurisdiction notes, but nothing about your situation suggests that acting on general information puts you at risk.
Yellow means your question stopped being general. You are describing your own dispute, your own contract, your own deadline, and the honest answer depends on facts an AI conversation may not fully capture. The answer still comes, in full, but with a recommendation to have an attorney review your specific situation before you act. Yellow is not an upsell trigger. It is the platform saying: this answer is a starting point, not a finish line.
Red means the question is safety-critical. The categories that draw red flags include active criminal exposure, deportation risk, custody emergencies, and injury claims that may be near or past a filing deadline. For these, the platform does not simply hand over an AI answer with a warning sticker. The answer is gated behind a human-attorney sign-off path, because in these situations a wrong or incomplete answer can cost someone their liberty, their immigration status, their child, or their claim. Red is deliberately rare, and it is the flag doing its most important work.
What should you do when you draw a yellow flag?
Treat the answer as preparation, not resolution. Read the reasoning, not just the conclusion, because the places where the explanation says "it depends" are the exact questions to bring to a lawyer. Then use the platform's paths to make the consultation cheap and fast: the practice-area workspaces help you organize your documents and facts, and Attorney Match connects you to a vetted attorney whose practice fits your matter, with your situation already framed. A yellow-flag user who arrives at a consultation with the AI answer, the jurisdiction note, and an organized set of facts spends the attorney's time on advice instead of reconstruction.
The worst response to a yellow flag is the most tempting one: taking the general answer, assuming your facts are typical, and acting. Most of the expensive legal messes attorneys see started as a situation someone reasonably believed was typical.
What happens when you draw a red flag?
You are routed toward a human. The answer path runs through attorney review rather than around it, and the platform's job shifts from explaining the law to getting you to the right help quickly: framing your matter, capturing the time-sensitive facts, and connecting you with an attorney or the appropriate emergency resource for your situation. If your matter involves a court date, an arrest, or a filing deadline, the speed of that handoff matters more than any paragraph of legal information, and the product is built to act like it.
It is worth saying plainly: a red flag is not the platform refusing to help. It is the platform helping in the only way that is actually responsible for that category of problem.
Why publish how the flags work?
Because a risk assessment you cannot inspect is just another black box. LawSensai provides legal information and self-help tools, not legal advice, and the flag system is where that line gets enforced question by question instead of buried in a footer. Knowing how the classification works also makes you a better user of it: mention your state, describe your posture honestly, and say so if there is a court date or deadline anywhere in the picture. The more the engine knows about where you actually stand, the more accurately it can tell you whether an answer is enough, or whether what you really need is a lawyer, and fast.