How NoVay works

NoVay is an AI, and it can be wrong

Replies are generated by a large language model. It predicts plausible text — it does not look things up, check facts, or know what happened recently. It can state wrong things with complete confidence, invent sources, quotes, statistics and citations, and make arithmetic mistakes. Treat everything as a draft to verify, not an answer to trust.

Not professional advice

NoVay is not a doctor, therapist, lawyer or financial adviser, and using it does not create any professional relationship. For medical, legal, financial or safety decisions, speak to a qualified person who knows your actual situation.

The safety layer, described honestly

Every request is combined server-side with a safety policy that user messages, pasted text and profile preferences cannot override. On top of that, the server runs a lightweight check on your latest message and, when it looks like a crisis or a request for dangerous instructions, adds an explicit instruction for how NoVay should respond.

NoVay declines to give actionable instructions for self-harm, violence, weapons construction or serious illegal and dangerous acts, and offers a safer alternative instead. Education, history, general safety information and fiction are allowed.

One content standard for everyone

NoVay applies the same conservative, youth-protective content standard to every account. There is no adult or explicit mode, and saying you are an adult, choosing Write or Business mode, roleplay or instructions in your message do not change it. NoVay won't write erotica or sexual roleplay, give explicit sexual coaching, or help with anything that could put a child at risk. Calm, non-graphic questions about puberty, bodies, consent, boundaries, healthy relationships and sexual health are welcome.

Before NoVay answers, an automated AI review checks your latest message. Replies are written in full on the server and reviewed again before they are shown, saved or available to read aloud — that is why text appears all at once rather than word by word. If the safety check itself is unavailable, NoVay shows nothing rather than an unreviewed reply. These reviews are run by an AI model provider and cost the app extra usage.

These safeguards reduce risks but aren't perfect. Automated checks can miss things or occasionally block something harmless, and no claim of complete safety is made here. Nobody monitors conversations around the clock; reports you send with “Child safety concern” are stored for the app operator to review, but no automatic escalation to authorities exists.

If you are in crisis

NoVay is not a crisis service. Nobody is monitoring your conversation, and no help is dispatched from this app. If you may be in immediate danger, contact your local emergency number right away and reach out to someone you trust. In the United States and Canada you can call or text 988 (Suicide & Crisis Lifeline). Elsewhere, please look up the crisis line or emergency number for your own country — NoVay does not know where you are and will not assume.

Limits you will see

Replies are capped per account per day and share a global daily budget, both enforced on the server. When a limit is reached, NoVay tells you plainly instead of failing quietly.