Focusing on person-agent fidelity, behavioural integrity, and dynamic digital permission slips, Daon’s new patent portfolio offers a comprehensive trust architecture for autonomous AI agents.
Digital identity trust organisation Daon has announced the issuance of its third major patent addressing the production challenges of agentic AI in highly regulated sectors. Issued by the U.S. Patent and Trademark Office on July 21, 2026, U.S. Patent No. 12,688,261 (Methods and Systems for Authorizing Invocation of a Tool by an Autonomous Artificial Intelligence Agent) allows enterprises to dynamically determine whether a specific AI agent may execute an action on a protected tool, service, or API.
Together with its predecessors, Daon’s three-patent portfolio establishes an end-to-end trust architecture spanning human-to-agent bonding, runtime behavioral monitoring, and real-time, action-level authorization.
As enterprises across financial services, healthcare, insurance, and government explore autonomous workflows, relying on traditional IAM (Identity and Access Management) built around static human logins falls short. AI agents plan, branch, parallelize, and execute tasks at machine speed, creating severe security vulnerabilities if granted blanket permissions.
“Agentic AI is unlikely to safely move from experimentation into high-value production environments on intelligence alone. It requires identity, policy, containment, and evidence at the moment an agent attempts to act,” said Tom Grissen, CEO of Daon and a co-inventor of the patent. “This patent strengthens Daon’s ability to help enterprises move beyond the false choice between blocking autonomous agents and granting them broad standing authority.”
The newly issued patent introduces an authorization checkpoint that generates a digital permission slip, encoding strict limits on an agent’s scope, permitted action type, and operational lifespan.
Daon’s complete three-patent control layer maps directly to the operational realities of enterprise deployment:
- Person-Agent Fidelity (U.S. Patent No. 12,452,035): Maintains the bond between a human and their agent using behavioral modalities and a Baseline Persona Model to detect drift and enable secure rollbacks.
- Agent Behavioral Integrity (U.S. Patent No. 12,563,045): Monitors runtime execution signals, applying policies at discovery, invocation, and runtime checkpoints to sandbox or deny risky behavior.
- Action-Level Delegated Authorization (U.S. Patent No. 12,688,261): Transforms fidelity, integrity, and request context into short-lived, narrowly scoped authorization artifacts at the exact moment of action.
“Traditional identity and access management was designed around a human authenticating and then operating within a session,” noted Ralph A. Rodriguez, President and Chief Product Officer of Daon, and first-named inventor across all three patents. “Autonomous agents can plan, branch, parallelize, and invoke tools at machine speed. This requires the control question to become more precise: Should this agent be allowed to perform this exact action, against this exact resource, right now?”
In sensitive environments where AI agents handle claims processing, document assembly, or account reviews, organizations must enforce tight boundaries. Daon’s architecture enables businesses to allow routine data analysis while requiring time-of-action verification for irreversible transactions or sensitive information releases.
With a global patent portfolio of more than 320 patents across biometrics, deepfake defence, and agentic AI, Daon is embedding this architecture into its future TrustX capabilities, giving enterprises the audit evidence and policy orchestration they need to scale AI safely.
