Now accepting early access

The question isn't
"Is this AI?"
It's "Can I trust this?"

TRST analyzes digital media across six layers of evidence and delivers a professional authenticity report in seconds — built for journalists, investigators, and legal professionals who can't afford to be wrong.

Free during beta · No credit card · Built for professionals

✓ You're on the list. We'll be in touch soon.

Or try it now → — no signup required.


Seeing is no longer believing.

For centuries, visual evidence was the gold standard. Today, a convincing deepfake video costs nothing and takes minutes to create. AI-generated images are indistinguishable to the human eye. Synthetic audio can clone anyone's voice from a 3-second clip.

The challenge is no longer whether fake content exists. The challenge is how you evaluate authenticity when you can't trust what you see — and the stakes are a published story, a court case, or an election.


Six layers of evidence.
One clear verdict.

TRST doesn't guess. It moves through a decision tree — each layer is a gate. If the evidence is conclusive early, we stop. If not, we go deeper.

Layer 1
C2PA Provenance
Checks for a cryptographic certificate embedded at the moment of capture. If valid, the chain of custody is intact and origin is confirmed.
✓ If verified — analysis complete
Layer 2
Metadata Forensics
Reads EXIF data — camera model, GPS, timestamp, software field. Stripped or inconsistent metadata is a strong risk indicator.
Layer 3
Origin Tracing
Reverse image and video search to find where this content first appeared online, who published it, and when. Misattribution is a form of misinformation.
Layer 4
Pixel & Signal Analysis
Runs the media through leading AI detection APIs that identify GAN fingerprints, compression anomalies, facial inconsistencies, and generation artifacts.
Layer 5
LLM Consensus
Multiple AI models independently reason over the evidence collected. Not a guess — a jury. All three must agree before high confidence is reported.
Layer 6
Context & Credibility
Evaluates source reputation, publication history, and surrounding context. A real image used deceptively is still misinformation.

A report you can actually use.

Not a score you have to explain. A professional document with evidence — shareable, exportable as PDF, and citable in editorial decisions, legal filings, and investigations.

TRST Authenticity Report
Analyzed June 14, 2026 · Image file
18
Trust Score / 100
C2PA
No cryptographic provenance certificate found
Not Found
Metadata
EXIF data stripped. Software field absent. No camera signature.
Stripped
Origin
First appearance: Telegram channel, 48 hours ago. No earlier source found.
Unverified
Detection
Sightengine: 91% probability AI-generated. GAN artifacts detected in background regions.
High Risk
LLM Consensus
Claude, GPT-4o, Gemini — all three independently flag as synthetic. High agreement.
Flagged
Context
Source account created 6 days ago. No publication history. Pattern consistent with influence operations.
Suspicious
⚠ High probability this content is synthetic or manipulated. TRST does not recommend publishing without independent verification from a primary source.

Built for professionals
who can't afford to be wrong.

📰
Investigative Journalists
Verify sources before publishing. Attach the TRST report to your editorial file.
⚖️
Legal Professionals
Generate a documented authenticity assessment for digital evidence in court.
🔍
Fact-Checkers
Go beyond reverse image search. Get a full forensic breakdown in seconds.
🛡️
Insurance Investigators
Detect fraudulent claim images and document the analysis for your file.
🏛️
Government & Policy
Identify influence operations and synthetic media campaigns at scale.
🎓
Researchers
Standardized authenticity framework for academic and institutional work.

We don't claim to detect every fake.
We help you decide what to trust.

Every AI detector will eventually be defeated by a better generator. That arms race has no finish line.

TRST is not a detector. It's a framework.

We combine provenance, forensics, detection signals, and expert reasoning into a single transparent report — and we show our work. Every finding is documented. Every signal is explained. No black boxes.

Our mission is to increase confidence in digital information without claiming certainty — because certainty is what gets you sued, and transparency is what gets you trusted.

Be among the first
to use TRST.

We're onboarding a small group of journalists, fact-checkers, and investigators for our private beta. Free access. Real feedback. Real impact.

No spam. No credit card. Just early access.

✓ You're on the list. We'll reach out within 48 hours.