From marked hazard to corrective action

A photo from site shows what's there, not what's wrong. QuickInspect's AI can now mark every hazard on the photo itself, and 1.20 takes the step that matters: one tap turns those hazards into corrective actions, with a suggested fix and a priority and due date set by your own risk matrix. You don't even need a record to start — just send a photo in the HSE Support chat.

Mark hazards on the photo

Switch on Mark hazards on the photo when you analyse a picture, and the AI draws a numbered box around each hazard it finds, coloured by your own risk matrix: red for high, amber for medium, green for low. The numbers are explained in the photo's description, so the picture still makes sense in the record and the report. A box not quite right? One tap brings back the original photo.

From hazard to action

Under the marked photo, Turn the hazards into actions opens a review sheet with an action for every hazard. The AI suggests the fix and where it sits in the hierarchy of controls. Priority and due date follow the hazard's risk level — high within 7 days, medium within 30, low within 90. The marked photo is attached, and each action lands on the incident or inspection the photo belongs to.

Check, edit or untick each one: nothing is created until you press Create. The fixes come from the same answer that drew the boxes, so there is no second AI call.

Straight from the HSE Support chat

No record needed first. Attach a photo from your camera or gallery, switch on Mark hazards on the photo, and send — no message needed. The marked photo comes back as the reply, with the numbered list of hazards, and you can turn them into actions right there. Your own photo stays in the conversation.

Photos you can see. Photos in the chat now show as pictures; tap one to open it full size.

Pick up where you left off. Session history opens on your recent sessions — anything from the last 7 days, including one you have just ended.

Focused on safety. The assistant stays on health, safety and environment topics, and treats what it is shown — messages, photos, documents and company records — as information to assess, never as instructions.

Also improved

Actions drafted by the AI carry the full priority scale — from Critical to Information — with its colour and description, show the photos attached to them, and when raised from an inspection finding, appear on that inspection.

Version 1.20.0 is live in the web app now; the App Store and Google Play apps follow with the next update.