How Nox works — from your message to a safe answer

Here is what actually happens when you talk to Nox — from the moment you send a message to the answer you read — including the safety layer that runs before the AI ever responds.

You ask in plain language

Open a chat and describe what's going on — a symptom, a medication question, a sleep problem, anything about your everyday health. No forms, no symptom checkboxes. You can type, talk out loud in a live voice call, or attach a photo of a label, meal, rash, or wearable reading.

Leo screens your message first

Before the AI model answers, Leo — Nox's medical-safety system — screens your message for signs of acute red-flag conditions across dozens of categories, such as stroke signs, chest pain, severe breathing difficulty, or a mental-health crisis. Its first check is an independent, deterministic detection layer that runs before any AI is called, surfacing clear guidance to seek appropriate care, with the correct local emergency number when you set your region. Leo is a safety net, not a guarantee: no automated system catches every emergency, and Nox publishes the deterministic detector's measured recall and false-positive rates on its Trust & Transparency page. This screening is deterministic — the same input always produces the same result — and it runs before any AI-generated text reaches you. When the deterministic layer finds nothing, a second check runs in the background: a lightweight AI classifier re-reads your recent messages to catch dangerous descriptions the fixed patterns can miss — slang, another language, older disease names, or indirect phrasing. If it recognizes a likely emergency, Nox surfaces the same seek-care guidance as a pattern match. This backstop can only add a safety note, never remove or soften one, and because it is not deterministic its results are kept separate from the published detector metrics. You choose how far Leo's screening goes, free on every plan, from a control right in the chat box. Three levels sit side by side, left to right, from least to most protective: Relaxed warns you about clear emergencies only; Standard also warns on anything urgent; and Strict adds the AI backstop that re-reads recent messages for dangerous descriptions worded indirectly or in another language. Whatever you pick, Leo always screens for true emergencies — the level only changes how many optional layers run on top — and Agent and voice conversations always use the strictest setting.

Nox answers clearly, citing trusted sources

If nothing urgent is detected, Nox explains what matters in plain language: likely explanations, what to watch for, sensible self-care, and when it makes sense to see a professional. Nox is instructed to ground health answers in trusted sources such as Mayo Clinic, CDC, NIH/MedlinePlus, and WHO — and never to diagnose or prescribe.

You can go deeper when you need to

Switch to the Thinking model for harder questions, or launch the Nox Agent (Pro and above) for deep research with cited, openable resource cards. Ask for a chart of your numbers, a diagram, or a structured 'Analyze My Day' report from your sleep, energy, and stress.

Nox remembers what you choose to share

With Memory on, Nox keeps track of durable facts you share — like an allergy or an ongoing condition — so you don't repeat yourself. You can review and delete every remembered fact in Settings, or use Incognito mode for conversations that are never saved at all.

You stay in control

Your chats are private, are not sold, and are not used to train OpenAI's models. App integrations (calendar, docs, tasks) only ever create new items and always ask you to confirm before acting. You can export questions to a real clinician, delete conversations, and manage your subscription anytime.