HAEVEN X
Search the guide
↑↓ navigate Enter open Esc close

Asking the AI hub for help

The AI assistant inside HaevenX can answer security questions, check the safety of a route, and summarize what just happened in your neighborhood.

4 min

Tap the brain-shaped icon at the top of the home screen and you're in the AI Security Hub. It's a chat interface to a security-tuned AI that knows your neighborhood, the recent incidents reported there, and the latest predictive heatmap. You can ask it questions in plain English.

9:41 HAEVEN X ● ● ●

Ai Hub

AI Security Hub chat screen with example questions

Tap the suggested prompts or type your own.

What you can ask

  • "Is it safe to walk home from school today?" — answers based on the predictive heatmap for your route at the current time of day.
  • "What happened in my neighborhood last night?" — summarizes any incidents reported in the last 24 hours.
  • "Show me the safest route from here to the library." — overlays a heatmap-aware route on the map.
  • "Are there any active emergencies near me?" — checks the live emergency feed for your area.
  • "What's the difference between a Silent Ping and a Visible Check-In?" — yes, it knows the app itself.

How it answers

Every answer the AI gives is based on real, current data. When it says "the area east of Main and 5th has elevated risk this evening", it's reading the same predictive heatmap you can see on the map screen. When it says "two incidents reported in your area in the last 12 hours", it's pulling from the same incident feed your forum shows. Every claim has a cite-able source.

Voice input

Tap the microphone icon next to the text box. Speak your question. It transcribes locally on your phone (no server-side audio) and sends the text to the AI. Useful when you're walking and don't want to type.

Daily limit

Each user can ask up to 10 questions per day. This is to keep the cost reasonable for the free tier and to encourage thoughtful questions over scrolling. Family masters get 25/day. There's no limit on reading the AI's daily summary that lands in your notifications.

What the AI is careful about

  • It will not name specific suspects or accuse anyone of a crime, even if a forum post does.
  • It will not give legal or medical advice; for those it tells you to call the right professional.
  • It will tell you when it doesn't know something, instead of guessing.