RBI's video-based customer identification is now a front line for deepfake and injection attacks. TrueSecurix adds a forensic deepfake layer to your V-CIP flow, catching AI-generated and injected faces that pass a liveness check, with evidence your compliance team can defend.
A pre-recorded or synthetic video fed into the camera pipeline to impersonate a live customer during V-CIP.
Fully synthetic applicants from diffusion and GAN models, with no real person behind the onboarding.
A victim's face mapped onto an attacker in the live session, the most common liveness bypass.
Send captured frames or the selfie to /v1/verify alongside your existing liveness. You get a deepfake risk score, a decision and the evidence, written to an immutable audit log, so every V-CIP decision is explainable and reviewable.
Read the API docs
curl https://truesecurix.com/v1/verify \
-H "X-API-Key: sk_live_..." \
-F selfie=@vcip_frame.jpg
# => "selfie.synthetic": "likely", "decision": "review"No. It is a forensic deepfake layer that runs alongside your existing V-CIP or liveness flow, adding detection of injected and AI-generated faces that liveness alone was not built to catch.
We provide an explainable, auditable deepfake risk signal that supports your V-CIP controls. We describe our posture as aligned with RBI KYC directions; we do not claim RBI empanelment or authorisation.
Every check writes an immutable record with the score, decision, evidence and the calibration version that produced it, so a reviewer can reconstruct any V-CIP outcome later.
100 verifications free, no card. Add a deepfake layer to your V-CIP flow.