One engine. Every identity. One trust score.
MFA-Brain™ is the decision engine at the centre of Authiun. It gathers identity signals from humans, devices and AI agents, weighs them together, and returns a score your journeys can act on in under 200 milliseconds.
Signal ensembling
Dozens of weak signals are combined into one weighted trust score rather than judged in isolation.
Policy thresholds
Each journey defines the score required to proceed, to step up, or to be blocked outright.
Zero-trust transport
Mutually authenticated channels and short-lived, scope-bound tokens for every session.
Explainable decisions
Every decision keeps the factors that produced it, so audit and fraud teams can trace outcomes.
Proof of person, gathered in the background
Authiun verifies the human without turning authentication into a chore. Signals are collected passively where possible and only escalate when the score demands it.
- Fingerprint and faceprint with anti-spoof liveness
- Trusted device binding with hardware-backed keys
- Geo, network and IP reputation context
- Behavioural typing and interaction patterns
Proof of provenance for autonomous software
An AI agent is only as safe as the identity behind it. Authiun binds each agent to an attested key, a registered scope and an observed behavioural baseline.
- Cryptographic attestation of the calling agent
- Registry lookup with declared, bounded scopes
- Behaviour baselining across call patterns
- Intent risk checks per requested action
Three layers of assurance
Each layer feeds the next. Together they keep trust current for the whole lifetime of a session.
Layer 01
Human identity
- Fingerprint and faceprint with anti-spoof liveness
- Trusted device binding with hardware-backed keys
- Geo, network and IP reputation context
- Behavioural typing and interaction patterns
Layer 02
AI agent identity
- Cryptographic attestation of the calling agent
- Registry lookup with declared, bounded scopes
- Behaviour baselining across call patterns
- Intent risk checks per requested action
Layer 03
Continuous session trust
- Re-scoring throughout the session, not just at login
- Automatic step-up when the score degrades
- Instant revocation of scoped agent tokens
- Full decision trail streamed to your SIEM
Watch MFA-Brain™ score a human and an AI agent
A step-by-step walkthrough of how Authiun evaluates identity signals in real time and builds a trust score before granting access.
MFA-Brain™
Risk score · person
Fingerprint match
Awaiting signal…
·Faceprint liveness
Awaiting signal…
·Device binding
Awaiting signal…
·Geo + network context
Awaiting signal…
·Behavioral typing pattern
Awaiting signal…
·
MFA-Brain™ trust score
Idle. Run the simulation
Authiun never grants access until the trust threshold is met.
Deploy without re-architecting
Authiun sits alongside your existing identity provider and services. Start with one journey, expand as trust in the engine grows.
Drop-in SDKs
Web, mobile and server SDKs with sensible defaults.
API-first
Score any request: login, transaction or agent call.
IdP friendly
Works with your existing directory and SSO estate.
Audit ready
Structured decision logs for compliance reporting.
See MFA-Brain™ on your journeys
See how Authiun scores humans and AI agents in real time, and what that means for your login, API and agent journeys.