Topic starter
02/08/2026 12:15 pm
TL;DR: Its MyFace passive liveness model achieved 0% APCER and 0% BPCER at both ISO 30107-3 Level 1 and Level 2 in iBeta testing, according to Yoti, while its facial age estimation received a smaller German buffer and showed 0.6% false positives for 13 to 17 year olds. The governance question is no longer whether these controls work in isolation, but how identity programmes balance assurance, user friction and policy choice across verification routes.
Editorial analysis by NHI Mgmt Group, based on content published by Yoti: “Thoughts from our CEO: Yotiâs latest MyFace model achieves a perfect score for Level 1 and Level 2 liveness detection”.
By the numbers:
- Yoti says 60% of 18 year olds and 80% of 19 year olds will pass the over 17.5 year check.
Key questions
Q: What breaks when age assurance relies only on a single verification method?
A: A single method can fail when users cannot complete it, when risk levels vary across sessions, or when borderline results need a stronger control.
Q: Why do digital identity programmes need explicit acceptance rules for choice?
A: Choice only works when every accepted route is mapped to the same assurance outcome.
Q: How do security teams evaluate whether liveness detection is strong enough?
A: Look for measurable resistance to presentation attacks, defined false accept and false reject rates, and testing that reflects the real environment where the control will run.
Practitioner guidance
- Define proofing route policy by transaction risk Map each verification journey to an acceptable route, such as passive liveness, active liveness, document proofing, or wallet-based credential presentation.
- Set age thresholds by measurable false-positive tolerance Choose the buffer and escalation path based on the business impact of underage access versus the cost of over-challenging adults.
- Standardise fallback routes for failed liveness checks Define what happens when passive liveness fails, including retry limits, switch to active liveness, and escalation to document verification.
Bottom line: The article shows that identity verification is moving toward route selection, where liveness, age estimation, and document proofing must be governed by transaction risk.
Explore further
View Full Forum → | NHI Foundation Course → | Our Services → | Read the full analysis →
02/08/2026 1:17 pm
Age assurance is becoming an assurance-routing problem, not a single-control problem. The article shows why organisations cannot treat liveness, age estimation, and document proofing as separate features. Each route creates a different balance of friction, fraud resistance, and decision confidence. The practical conclusion is that IAM teams need a governance model for selecting the right proofing route by risk, not by convenience alone.
A question worth separating out:
Q: What should organisations do when facial age estimation is close to the policy threshold?
A: They should treat the threshold as a policy decision, not a model artefact. If a result falls near the cutoff, the relying party should define whether to challenge the user, accept a lower-risk action, or require stronger proof. That keeps the decision consistent and defensible across transactions.
👉 Read our full editorial: Digital identity verification now hinges on liveness, age checks and choice