Identity proves who
Passwords, identity verification, biometrics, and KYC can establish identity.
An AI-era trust startup
Proving human intent behind critical actions.
AI agents can initiate actions. HumanProof verifies when an authorized human intentionally approves them.
HumanProof creates cryptographic evidence linking human confirmation to the exact action being authorized—helping organizations distinguish legitimate human intent from autonomous AI, bots, replay attacks, and manipulated sessions.
- Multi-signal
- Verification approach
- Action-bound
- Intent confirmation
- Audit-ready
- Evidence records
- Startup
- Pre-seed, building
Verification path
- AI Agent
- Critical Action
- HumanProof
- Human Confirmation
- Verified Action
The HumanProof node is the trust layer sitting between automated systems and consequential actions.
The problem
Identity is no longer enough.
As AI agents gain the ability to initiate payments, change accounts, execute transactions, approve workflows, and interact with critical systems, traditional authentication leaves an important question unanswered.
Devices prove possession
Passkeys and trusted devices can establish control of a credential or device.
Neither necessarily proves intent
A valid identity or authenticated device does not automatically prove that a human intentionally authorized the exact action being executed.
HumanProof adds the missing layer: verifiable human intent.
The solution
A human confirmation layer for the AI economy.
HumanProof evaluates multiple signals surrounding an action before producing a single human-confidence decision.
- 01
Action Requested
A sensitive action enters HumanProof.
- 02
Evidence Collected
Multiple human and contextual signals are evaluated around the action.
- 03
Intent Confirmed
The user completes an intentional challenge tied to the exact action.
- 04
Risk Analyzed
Risk and integrity are assessed across the confirmation and its surrounding context.
- 05
Decision Generated
A unified human-confidence outcome is returned.
HUMAN CONFIRMEDREVIEWREJECT - 06
Proof Recorded
A tamper-resistant verification record can be stored for audit and investigation.
Human confirmation engine
Multiple signals. One human-confidence decision.
Human Presence
Presence signals help confirm that a real person—not an automated system—is participating in the action.
Authorized Identity
We build on the identity, device, and authorization checks your systems already perform, rather than replacing them.
Action-Bound Intent
Every confirmation is tied to the exact action being authorized, so approval can't be replayed against a different transaction.
Risk & Integrity
HumanProof evaluates risk and integrity across the interaction to flag suspicious or manipulated approval attempts.
Higher-assurance deployments may incorporate additional biometric or sensor-based evidence where appropriate.
What makes HumanProof different
HumanProof complements authentication—it does not replace it.
HumanProof combines human confirmation, intentional action binding, risk analysis, and evidence generation rather than replacing existing identity systems.
| Capability | Traditional CAPTCHA | Identity Verification | Authentication / Passkeys | HumanProof |
|---|---|---|---|---|
| Detect automated interaction | — | — | ||
| Verify identity | — | |||
| Verify device possession | — | — | ||
| Verify human presence | — | |||
| Verify intentional approval | — | — | — | |
| Bind approval to exact action | — | — | — | |
| Detect replay attempts | — | — | ||
| Generate audit evidence | — | — |
Use cases
Built for actions where intent matters.
AI Agent Payments
Require human confirmation before an autonomous agent completes a high-value payment.
Banking & Fintech
Verify intentional approval for transfers, withdrawals, account changes, and sensitive transactions.
Enterprise AI
Place HumanProof between AI agents and consequential enterprise actions.
Account Recovery
Add stronger confirmation before credential resets or security-critical account changes.
Healthcare & Sensitive Records
Verify human authorization before releasing or changing protected information.
Government & Critical Infrastructure
Provide additional evidence for high-risk digital approvals and administrative actions.
Developer platform
One trust layer. Simple integration.
Applications send HumanProof evidence surrounding an action. HumanProof evaluates the signals and returns a unified human-confidence decision.
Example request
POST /v1/verify
{
"action": {
"type": "payment",
"amount": 25000,
"currency": "USD"
},
"confirmation_token": "hp_conf_a1b2c3d4"
}Example response
{
"decision": "human_confirmed",
"confidence": "high",
"evidence_id": "ev_4f9c21a7"
}Architecture
Designed to sit between intelligence and action.
User / AI Agent
Application
HumanProof Trust Layer
A focused verification layer that sits between your application and the consequential action it wants to authorize.
01Capture signals
02Human-confidence decision
03Tamper-resistant evidence
HUMAN CONFIRMED
REVIEW
REJECT
Bank / Payment / Enterprise System / API / Critical Action
Why now
AI is changing the meaning of authorization.
For decades, digital security assumed that an authenticated account was being operated directly by a person.
AI agents change that assumption.
An authenticated user may increasingly delegate actions to software capable of operating continuously and autonomously.
The next generation of trust infrastructure must therefore distinguish between authenticated access and intentional human authorization.
Yesterday
Authenticated access
A valid session was assumed to mean a person.
Now
Intentional human authorization
Evidence that a human meant to approve this exact action.
HumanProof is building infrastructure for that transition.
Current technology
Already beyond the concept stage.
HumanProof is an active prototype and MVP under continuous development and testing.
PrototypeMVPTestingIn development
Trust-check API
Human approval workflow
Persistent verification records
Human-confidence scoring
Liveness verification
Action binding
Replay protection
Automation-risk scoring
Adversarial verification testing
API-based integration
Audit-ready verification records
Vision
The trust layer between humans and autonomous systems.
HumanProof begins with verifying intentional human approval.
The larger opportunity is a universal trust infrastructure through which organizations can establish when humans, AI agents, devices, and automated systems are authorized to act—and create verifiable evidence of how those decisions were made.
In an internet filled with intelligent machines, proof of human intent becomes infrastructure.
About
A startup proving human intent behind critical actions.
HumanProof is an early-stage startup building trust infrastructure for the AI era. Iskender Pasquali is its Founder and CEO. With a background in marketing, he has spent his career understanding how trust is built between brands and the people they serve—and how easily that trust can break down when the experience falls short.
He recognized early that the rapid adoption of AI agents would fundamentally destabilize a core assumption of digital trust: that an authenticated session is always operated by a human. He founded HumanProof to address a growing challenge: proving that critical digital actions are intentionally authorized by real humans, not autonomous AI agents.
Drawing on his marketing perspective, Iskender sees trust not as a technical feature but as the foundation of every meaningful digital relationship. Under his leadership, HumanProof is building the infrastructure that lets organizations distinguish between an authenticated account and an intentional human action.
The mission is simple: make proof of human intent a first-class part of digital trust.
Get started
Ready to verify human intent?
Explore how HumanProof can protect critical actions in an AI-driven world.
Contact us: hello@humanproof.io