Loading…
Trust & security
How Gramavec measures, what it stores, and the line we hold: an assistant, not a verdict.
Gramavec's forensics engine — the signature and document measurement — is classical computer vision that runs in-process. For the measurement itself there is no external AI model, no network call, and no database: the same input always produces the same output, with reproducible visual evidence.
Analyses are retained according to your plan — Free for 24 hours, Professional for 90 days, Business for 1 year, and Enterprise as configured — then removed. You can delete a report at any time.
Working in a regulated industry — banking, notarial, legal? Enterprise plans can discuss data-residency and deployment requirements. Tell us your constraints and we'll work through them with you.
For documents and signatures, Gramavec is a pre-screening assistant: it flags what deserves a closer look and shows the measured evidence. A certified Forensic Document Examiner keeps the final say — we never claim an authenticity verdict.
Thresholds are derived from the public CEDAR signature dataset. The numbers below are what the engine reports — no marketing inflation, no hidden tuning.
0.72
Forensic AUC
0.37
Equal error rate
0.77
FRR at FAR 1%
60+40
Genuine + impostor
CEDAR is the public Center of Excellence for Document Analysis benchmark; 60 genuine + 40 impostor (n=100) is a small sample. We report all four metrics including the worst-case (FRR at FAR=1% = 0.77) so reviewers can audit the gap between best- and worst-case performance. Source: gramavec/forensics/calibration_cedar.json. Pinned by automated tests so a calibration regression cannot ship without a code change.
MethodologyOur handling of personal data is set out in the policies below.
For enterprise security reviews, data-residency, or deployment questions, get in touch.
Contact us