The authenticity authority for digital evidence.
CanonProof is the neutral trust layer for the internet β cryptographically certifying files and exposing explainable forensic reasons behind every result.
Neutral Trust Authority
Standards-driven verification, ES256-signed certificates, public JWKS β the SSL-like layer for evidence authenticity.
Explainable Forensics
Indicators for metadata, ELA, DCT/entropy, thumbnails, headers, and AI-likelihood β with human-readable reasons you can cite.
Open Verification
Every certificate has a public verification URL and signed JWS payload for independent validation anywhere.
How CanonProof certifies authenticity
Upload & identify
Image, video, audio, or PDF. CanonProof hashes with SHA-256 and identifies media type.
Extract indicators
EXIF plausibility, thumbnail v. image pHash, ELA residuals, header heuristics, and AI-likelihood signals.
Score & label
Weighted scoring produces one of four labels with an overall confidence.
Sign & publish
Results are ES256-signed (JWS) and issued with a certificate ID and a public verification URL.
Explainable reasons (example)
Human-readable lines that map indicators to the final label β built for filings, reports, and audits.
- img.exif.has_make_model: present - img.thumb_mismatch: false - img.ela.p95: 0.016 (low residuals) - ai.likelihood: 0.11 (low) β LIKELY_AUTHENTIC (score 92, confidence 0.94)
Who relies on CanonProof
API-first by design
POST /api/v1/verify β signed JWS, score, reasons, and a public verify URL.