TRUSTLENS
Sign in
Explainable synthetic media detection

Don't ask if it's fake.
Ask if you can trust it.

TrustLens analyzes images, video, and audio for manipulation signals, then shows its work — a trust score, flagged regions, and a plain-English explanation, not just a verdict.

See how it works

3

Media types

SHA-256

File fingerprint

JSON

Evidence report

Open Innovation · AI & emerging technologies

Synthetic media is easy to make and hard to verify.

Who is affected

Everyday people, students, journalists and small organisations who receive images, videos and voice notes they have no way to verify before trusting or forwarding them.

Why current tools fall short

Most detectors return a bare "real" or "fake" label. Without reasons, a verdict is hard to trust, hard to challenge and impossible to hand to someone else as evidence.

What TrustLens does differently

It reports a trust score with plain-English reasons, flags what looked inconsistent, fingerprints the file with SHA-256, and packages everything into a report anyone can review.

How it works

From upload to evidence in four steps.

  1. 01

    Upload

    Add an image, video or audio file.

  2. 02

    Analyse

    The file is fingerprinted and checked for manipulation signals.

  3. 03

    Explain

    You get a trust score, flagged findings and a plain-English summary.

  4. 04

    Share

    Download the evidence report so others can verify your result.

Evidence Locker

Results you can hand to someone else.

Every scan is saved to your history. Each report includes the file's SHA-256 fingerprint, the trust score, the verdict and every finding, so the result can be checked independently.

Prototype note: without a detection API key, TrustLens runs in a clearly labelled demo mode that simulates the score. Add a provider key to switch to live detection.