KYC and identity document fraud detection

Onboarding fraud increasingly uses edited ID documents and AI-generated or face-swapped selfies. sealverity.ai checks the ID front and back together with a selfie photo or video and returns a clear Pass, Review or Reject decision card that an analyst can act on.

Document checks

The ID images are scored by the detector ensemble and reviewed visually by AI. The machine-readable zone (MRZ) is parsed and its check digits validated, and the printed fields are cross-checked against the MRZ using OCR. EXIF data and compression artefacts are inspected for signs of editing.

Duplicate detection compares each document against earlier submissions using perceptual hashing and face comparison, which helps catch the same document or face being reused across different applications.

MRZ parsing with checksum validation
OCR cross-check of printed fields
EXIF and compression analysis
Duplicate detection by perceptual hash and face

Selfie and liveness checks

The selfie is scored for deepfake and face-swap indicators and compared with the photo on the ID to produce a face-match similarity. For selfie videos, replay indicators are checked to spot a screen being filmed instead of a live person.

Pass, Review or Reject

All signals are combined into a decision card with configurable thresholds, so each organisation decides how strict it wants to be. Borderline results go to Review rather than being forced into Pass or Reject, and analysts record the final decision with a reason.

API for onboarding flows

Identity checks can be submitted through a REST endpoint as multipart form data with an organisation API key, with an optional webhook callback when the result is ready. This lets onboarding systems send checks automatically while analysts handle the review queue in the app.

Privacy by default

ID images are stored privately and deleted automatically under the organisation's retention policy, 30 days by default, unless they are attached to a case. Personal fields are masked for everyone below supervisor level. Every view and decision is recorded in the hash-chained audit log.

Part of one investigation platform

This capability is part of sealverity.ai, a single platform for media authenticity and investigations. Findings from images, video, audio, live calls, identity checks and social and news monitoring can be linked to cases, entities and a link graph, with risk scores that show which factors contributed. Access is protected by mandatory multi-factor authentication, single sign-on and SCIM provisioning, five roles, retention policies and legal hold. Organisations can bring their own self-hosted detectors or switch on sovereign mode, which blocks every external call on the server and logs each blocked attempt. Integrations include a browser extension, Slack and Teams, email scanning, webhooks, SIEM connectors and a REST API.