●Victim support

If a fake image of you is being shared.

You are not to blame. Witness can help you document that an image or video shows signs of manipulation, so you can report it to platforms and, if needed, the police.

Multi-model detection<3s per imageZero retention
reported_image.jpgscan complete
Submitted evidence
AI-generated
0.94LIKELIHOOD
Save report
Face-swap boundary
91
Generative fingerprint
84
Frequency artefacts
78
C2PA signature
0
<3sMedian result time
4Independent signal families
0Files retained
C2PAProvenance read

●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.

reported_image.jpgintake · 10:42:08
AI-generated
reported_image.jpgintake · 10:42:11
Real
reported_image.jpgintake · 10:42:15
AI-generated

01

Detect

Every file is scanned by independent models for images, video and faces, in the web tool, the extension or by API.

Face-swap boundary
91
Generative fingerprint
84
Frequency artefacts
78
C2PA signature
0

3 / 4 signals corroborateCalibrated likelihood

02

Explain

See which signals fired and how strongly. No single model decides; every result comes with its reasons.

Document
Report
Get help

Rule: likelihood ≥ 0.85 → Report<3s

03

Route

Branch your workflow on the result. Clean items pass, flags go to review with a full audit trail.

●What we catch

What you can do now.

Flagged at the point of submission, before it reaches a reviewer.

01

Don’t engage

Don’t reply to the person sharing it, and don’t pay anyone to remove it.

02

Document it

Save links, usernames, dates and screenshots. Scan the image and download the result.

03

Report it

Use the platform’s report tools, most remove non-consensual and manipulated intimate images.

04

Get support

Talk to someone you trust. If it’s sexual, threatening or involves a minor, contact the police.

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.

Live

Video

Face swaps across 32 frames and fully generated video from Runway, Kling, Veo and Pika.

Live

Voice

Cloned and synthesised speech in calls, notes and recordings.

Live

What you get

Each file is marked Document, Report or Get help, with the likelihood, signals and audit ID behind it.

One answer

●Next steps

You don’t have to handle it alone.

Include the link and say the image is manipulated or non-consensual.

Services like StopNCII.org can help stop intimate images being re-shared on partner platforms.

Creating or sharing intimate deepfakes is a crime in many countries.

●How it works

Three steps. One result.

01

Scan the image

Upload it privately. Media is analysed and not stored.

02

Save the result

Download the result and scan ID for your records.

03

Report it

Attach the result when you report to the platform or police.

●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

Available

Upload in the web app. Images, video and voice, up to 50 MB. Full results in seconds.

witness.vision/scan
Page media● Scan
reported_image.jpg · AI-generated · 0.94

Chrome extension

Available

Scan any image or video from inside the tools your team already uses. No download, no upload.

portal.example.com
Page media● Scan
reported_image.jpg · AI-generated · 0.94

API

Soon

Single scans, batch jobs and webhooks. Structured JSON with result, likelihood, signals and audit ID.

POST /v1/detect
curl -X POST https://witness.vision/api/v1/detect \
  -H "Authorization: Bearer $WITNESS_KEY" \
  -H "Content-Type: application/json" \
  -d '{"media_url": "https://…/reported_image.jpg"}'

→ 200 OK · 412ms
result: ai_generated · 0.94

●Why Witness

Built to be checked.

We say what the system does, what it returns and where its limits are.

Specialized models

Detection built per signal

Specialized models per signal, so you get more coverage and fewer false positives than one classifier.

Explainable

Reasons, not just a score

Every result shows the signals behind it and where in the file they fired.

Zero retention

Analysed, then gone

Media is never stored, shared or used for training.

C2PA

Provenance, when it exists

Reads Content Credentials from capture device to current file.

●FAQ

Common questions.

Something else? Talk to our team

Will Witness store the image?

No. Media is analysed and discarded.

Can Witness take the image down?

No. Only the platform hosting it can remove it, but a scan result can support your report.

Is a Witness result proof?

It’s a strong signal, not certified forensic evidence. For legal proceedings, ask for a qualified digital forensics expert.