●Insurance · Claims & SIU
Catch AI-generated claims evidence at intake.
Screen damage photos and injury documentation for AI generation before they reach your adjusters. A result, a likelihood and the signals behind it, in seconds.
●Platform
Detect it. Explain it. Act on it.
Every scan returns the same three things: a result, the signals that produced it, and a decision you can route on.
01
Detect
Every file is scanned by independent models for images, video and faces, in the web tool, the extension or by API.
3 / 4 signals corroborateCalibrated likelihood
02
Explain
See which signals fired and how strongly. No single model decides; every result comes with its reasons.
Rule: likelihood ≥ 0.85 → Flag<3s
03
Route
Branch your workflow on the result. Clean items pass, flags go to review with a full audit trail.
●The problem
Fake damage photos are getting paid.
Generative tools produce photorealistic vehicle, property and injury imagery in seconds. Adjusters reviewing hundreds of claims a week have no reliable way to separate it from real evidence at scale.
●What we catch
What gets flagged.
Flagged at the point of submission, before it reaches a reviewer.
01
AI-generated vehicle damage
Fabricated dents, crash scenes and collision damage from image generators.
02
Fake property damage
Storm, fire and water damage imagery that never happened.
03
Staged accident images
Composites and edits that build a scene from parts.
04
Recycled claim photos
The same source image reused across claims or accounts, including edited variants.
05
Manipulated injury records
Altered or generated medical imagery in workers comp and liability claims.
06
Face swaps
Impersonated policyholders or fabricated witnesses in claims documentation.
The pipeline
A detector for each kind of media.
Image, video and voice each have their own detector. Their findings come back as one clear answer you can act on.
Image
Diffusion, GAN and face-swap images from Midjourney, DALL·E, Stable Diffusion, Flux and Firefly.
LiveVideo
Face swaps across 32 frames and fully generated video from Runway, Kling, Veo and Pika.
LiveVoice
Cloned and synthesised speech in calls, notes and recordings.
LiveWhat you get
Each file is marked Pass, Flag or Review, with the likelihood, signals and audit ID behind it.
One answer●Lines of business
Every line you write.
Screen crash photos, collision damage and staged accident imagery before the claim is processed.
Verify storm, fire and water damage photos in homeowner and commercial property claims.
Screen injury documentation and medical imagery for AI generation or manipulation.
Verify equipment damage, inventory loss and business-interruption evidence.
●How it works
Three steps. One result.
Submit evidence
Upload in the web tool, scan in the claims portal with the extension, or send by API.
Ensemble analysis
Pixel forensics, frequency analysis, face-swap detection and provenance run in parallel.
Result & route
Pass, flag or review, with a likelihood and the signals that drove it. Under 3 seconds per image.
●Integration
Three ways in.
Start with the web app.
Scan in the web app or Chrome extension today. The REST API and MCP server for your AI agents are coming soon. Same detectors, same result, same audit trail.
Web app
AvailableUpload in the web app. Images, video and voice, up to 50 MB. Full results in seconds.
Chrome extension
AvailableScan any image or video from inside the tools your team already uses. No download, no upload.
API
SoonSingle scans, batch jobs and webhooks. Structured JSON with result, likelihood, signals and audit ID.
curl -X POST https://witness.vision/api/v1/detect \ -H "Authorization: Bearer $WITNESS_KEY" \ -H "Content-Type: application/json" \ -d '{"media_url": "https://…/auto_claim_48213_rear.jpg"}' → 200 OK · 412ms result: ai_generated · 0.93
●Why Witness
Built to be checked.
We say what the system does, what it returns and where its limits are.
Detection built per signal
Specialized models per signal, so you get more coverage and fewer false positives than one classifier.
Reasons, not just a score
Every result shows the signals behind it and where in the file they fired.
Analysed, then gone
Media is never stored, shared or used for training.
Provenance, when it exists
Reads Content Credentials from capture device to current file.
●Pricing
Priced around your volume.
No minimums and no long-term contracts. Tell us your volume and we will quote within one business day.
Free
$0
- 20 image scans a month
- 2 min video/voice total
- Full heatmaps
- Generator attribution
- EXIF/C2PA/SynthID
- Web and Chrome extension
●Pro
$15 / month
- 1,000 image scans a month
- 15 min video + 10 min voice
- Everything in Free
- Faces and face-swap specialist
- Scan by URL: paste any link
- Email support
Enterprise
Custom
- Volume pricing
- Everything in Pro
- SSO and audit logs
- Dedicated support
- Custom SLA
- On-premise option
●Explore
One engine. Every surface.
The same detection, result and audit trail across every team that handles media.
Detection
AI image detector
Find out if an image was made by AI.
Explore →Detection
Face swap detection
Find out if the face in a photo or video has been replaced, reenacted or lip-synced.
Explore →Detection
C2PA checker
Read the Content Credentials inside an image or video: who signed it, what captured it, and every edit since.
Explore →Detection
AI video detector
Find out if a video was generated by AI or contains a face swap.
Explore →Product
Chrome extension
Scan media on any site, social feeds, news, claims portals, case files, and get a result in the side panel..
Explore →How does AI-generated insurance fraud work?
Generative tools create realistic photos of damage or injuries that never occurred. These are submitted as claims evidence to trigger payouts for fabricated losses.
What claims can be screened?
Auto, property, workers compensation and commercial lines, any claim where photos or video are submitted as evidence.
Does Witness store claim photos?
No. Images are analysed and discarded. No claim photos, policyholder data or claims metadata is stored, retained or used for training.
How accurate is detection?
Accuracy varies by generator and media type. Every scan returns a likelihood and the signals behind it, so reviewers can see why a result was reached.