Multiple evidence layers
A valid identity reference, a matching face and an authentic capture path answer different questions. The engine brings those signals into one verification decision.
CONNECTED SECURITY DECISIONS
Every signal.
One considered decision.
FaceX’s risk engine brings deepfake detection, camera authenticity, anti-spoofing and identity evidence together. It supports approval, review or rejection based on the combined picture, rather than a single reassuring score.
Evidence combined
Media, camera, presence and identity.
Critical risks retained
Integrity flags remain decision evidence.
Clear outcomes
Approve, review or reject.
THE PROTECTION LAYER
Risk engine, explained.
Critical integrity failures
An injected feed, synthetic media or replay evidence can remain a blocking concern despite a strong face match. The illustrative policy makes that relationship explicit.
Uncertainty handled
Insufficient evidence should remain visible as uncertainty. The engine can route a case for review or another capture instead of treating incomplete checks as success.
FROM CAPTURE TO CONFIDENCE
How the evidence connects.
The decision layer combines the required evidence and the configured policy. The examples use simple qualitative outcomes; they do not expose production thresholds or weights.
THE EVIDENCE JOURNEYConceptual security flow
Media & presenceDeepfake and anti-spoofing evidence
Camera integrityInjection and virtual feed checks
Identity evidenceReference and face comparison
FaceX risk engineCombine evidence with the decision policy
Policy outcomeRequired checks and unresolved risks considered together
ApproveRequired evidence is consistent
ReviewMore evidence or attention needed
RejectBlocking risk or failed requirement
SEE THE DIFFERENCE
Evidence changes the outcome.
Choose a scenario to see how different signals can affect an example decision.
Illustrative scenarios only. These qualitative signals and policies do not expose production scores or thresholds. No camera access, biometric capture or live check occurs on this page.
EVIDENCE MATTERS
Confidence with context.
The security checks answer different questions. Complete identity verification needs their combined evidence.
Uncertainty remains visible
An inconclusive check is not the same as a passed check. Review and recapture paths can preserve that distinction in the applicant journey.
Performance has a scope
FaceX’s reported 99% deepfake figure does not establish a separate camera-integrity, anti-spoofing or full-system success rate. Read the performance context.
COMMON QUESTIONS
A clearer understanding.
Does a high face match override a security failure?
In the illustrative policy, no. A face can resemble the identity reference while the media, capture source or presentation remains untrusted.
What is the difference between review and rejection?
Review represents unresolved or insufficient evidence. Rejection represents a failed requirement or blocking risk under the decision policy. Production routing depends on the published integration and configured policy.
Are these the production API fields?
No. The developer reference presents proposed contracts and illustrative evidence summaries. Confirm actual fields, decision semantics and thresholds against the live integration contract.
Further reading: NIST guidance on digital injection and forged media and NIST presentation attack detection evaluation.These sources explain the threat categories and evaluation concepts. They do not certify FaceX or validate its reported performance.