Product · Proof of humanhood
Proof of humanhood
Verify a real person is present, and keep the evidence.
Proof of Humanhood is Plurall AI's verification surface for onboarding and high-risk actions. It combines multi-factor liveness with deepfake and injection detection on the live camera feed, then records what it saw, so a reviewer, an auditor or a regulator can see why a check passed or failed.
Built for the moment a fake identity tries to get in
- Marketplaces
- Verify sellers and high-value buyers before they list or transact, and cut synthetic accounts at signup.
- Financial services onboarding
- Add deepfake-resistant liveness to KYC flows without replacing your existing identity vendor.
- Social and community platforms
- Confirm humanhood at signup and on account recovery, where impersonation starts.
- Trust and compliance teams
- Keep an audit trail of every verification, with the evidence behind each outcome.
How it works
- Capture. The user completes a short liveness prompt on a live camera feed.
- Check. Sage, Plurall AI's live-camera model, performs real-time biometric verification: presentation-attack detection, injected-content detection and multi-factor liveness, while the session is happening.
- Decide with evidence. You receive a liveness status with the evidence behind it, and every check is written to an audit trail. Borderline sessions are routed to review rather than guessed.
What you get
| Capability | What it means for you |
|---|---|
| Liveness status | Pass, fail or review, with the reasons attached |
| Spoof and injection detection | Presentation attacks and injected synthetic content caught on the live feed |
| Evidence, not a score | Reviewers see what fired and why, so decisions can be defended and appealed |
| Audit trail | Every verification recorded with its outcome for compliance reviews |
| Case escalation | Suspicious sessions route into Plurall AI fraud defense for investigation |
| API and workflow integration | Call it from your onboarding flow or review tooling through the Plurall AI API |
Why evidence matters
A score tells you a check passed. Evidence tells you why, which is what an auditor asks, what a user appeal needs and what a regulator expects under emerging synthetic media rules in the US, EU and UK. Plurall AI returns both.
Frequently asked questions
What is proof of humanhood?
Proof of humanhood is a verification step that confirms a real, live person is present at the moment of a check, rather than a replayed video, a generated face or an injected stream. Plurall AI pairs it with deepfake detection and an audit trail.
How is this different from liveness detection?
Liveness detection asks whether the capture shows a live person. Proof of Humanhood adds deepfake and injection checks on the same session and records the evidence, so the result holds up in review.
Does it replace my KYC or identity verification vendor?
No. It adds a deepfake-resistant liveness layer to your existing identity flow through the API.
What happens when a check fails?
You get the evidence and a liveness status. Failed or uncertain checks can be escalated into Plurall AI fraud defense for human review.
Sage, the live-camera model behind Proof of Humanhood, is described on the engineering page. Developer docs: plurall.tech/docs.