Yubico, IBM and Auth0 partner on human-in-the-loop agentic AI security

Cybersecurity firm Yubico has entered into a collaboration with IBM and Auth0, introducing architecture comprised of a human-in-the-loop model that enables AI systems to operate autonomously for routine tasks, while escalating high-risk actions for secure human approval.

At the centre of the solution is Yubico’s hardware-backed authentication delivered through the YubiKey, a collection of hardware authentication devices made by the company. This ensures approvals for sensitive actions are tied to a verified individual and physical presence.

When a high-risk action is triggered, Auth0 initiates a secure approval request via backchannel authentication, while IBM’s AI agents orchestrate the workflow.

The authorised user must physically authenticate using a YubiKey, providing cryptographic proof that the correct individual approved the action, the approval cannot be replayed or intercepted and the decision is securely bound to a verified identity.

The collaboration brings together IBM’s WatsonX AI capabilities to analyse data and execute workflows, Auth0’s identity orchestration to manage secure approval flows and Yubico’s hardware root of trust to enforce human verification.