iProov launches human authorization for AI agent gateways

Identity verification company iProov has launched a Human Authorization Service built for AI agent gateways, and said the product already has its first integrations. The framing is deliberate: the goal is not to make agents more capable, but to require a verifiable human permission before an agent takes a consequential action.

The gateway is the natural place for that check. It sits between the agent and the systems it touches, handling identity, permissions, routing and audit. When an agent is about to move money, change a configuration, reach sensitive data or commit a user to something externally, an API token alone cannot show that a real person agreed to this particular step.

From what has been disclosed, the service is already being integrated, though the announcement does not detail the shape of those first deployments or name the partners involved. What is clear is the architectural choice: authorization lives at the gateway rather than being reimplemented inside every agent application, which makes it far easier to roll out at scale.

The launch points at a widening gap in agent deployments. Authentication answers who someone is; authorization answers what they may do. When the decision itself is generated by a model in real time, it is often unclear who owns that specific step, and making human approval an explicit, logged action gives automated workflows an accountability anchor.

For enterprises, the value is less about technical novelty than about whether the control survives an audit. Regulators and internal risk teams tend to ask three things: was a human involved for sensitive actions, can the record be reconstructed later, and does the approval flow slow the product down.

What to watch next: the integration list, which agent gateways and protocols are supported, the latency and failure rate of the approval step, and how the service connects to existing identity stacks such as single sign-on, permission management and ticketing-based approvals. Those details decide whether human authorization becomes standard agent infrastructure or stays a compliance talking point.