Help

What does AI photo review do?

By the LockProof Team · Last updated November 11, 2026

AI photo review is an optional pass that marks a scheduled photo as plausible, unclear or suspicious, with a confidence, a note and a reason. It is an advisory flag for your office, never a verdict. It never changes a check’s status and never blocks a submission.

← All help articles

When it runs and what it reads

It runs after the photo is saved and the worker’s submission has gone through.

It runs only when LockProof’s AI service is set up, your plan includes AI review, and the setting is switched on for your account.

It asks whether the photo looks plausible, based only on the image. It does not certify the scene.

This page covers the review on scheduled photo requests.

If the AI layer writes the dashboard’s Compliance Brief, it reads “AI summary · advisory”.

Common questions

Can AI review reject a photo?

No. It never blocks a submission and never changes the status. It adds a flag the office can read.

Does a flag mean the photo is fake?

No. A flag is a prompt for a person to look. The counts come from the records, not from the flag.

A film in six steps

Watch what AI review does

One photo, from the camera to the advisory mark, and how to search what the AI wrote. Use the chapter buttons to jump to a step.

This film is an illustration. The people, buildings and companies in it are invented, and so are the photos, messages and numbers on its screens. The screens are drawn to follow the app. The labels and pictures around them are the film’s.

All films

Read the transcript

Opening 0:00

Two hundred photos came in this week. Nobody can open every one. AI review flags the ones worth a second look.

The photo 0:08

The photo arrives the usual way. Live from the camera, through the link LockProof texted. It is sealed when it arrives. Our server sets the time, and a fingerprint covers the record. The location comes from the worker’s phone when they share it, and you can require it.

The review 0:27

Then the AI looks at the photo. It marks it plausible, unclear, or suspicious. It is told to flag things like a black screen, a blur, or a photo of another screen.

The search 0:38

You can search what the AI wrote. Say you remember broken glass in a photo, but not which building or which day. Type it, and the list narrows to the records whose AI text contains it.

The reason 0:51

Open a record to read its reasoning. The review is advisory. It never changes a record’s status, and it never blocks a photo.

Your rules 1:00

You decide what a good photo should show. Or keep the default for your industry.

The brief 1:06

The dashboard opens with a written brief. It is an advisory summary. The counts under it come from the records. LockProof. Know the work got done, without calling anyone. See it for your own team at lockproof.com.

See it on your own sites