WideField Security wins two patents for identity threat detection tech

By AI, Created 5:01 PM UTC, May 26, 2026, /AGP/ – WideField Security said June 2, 2026, that the U.S. Patent and Trademark Office granted two patents for its identity threat detection and posture management technology. The patents cover live session graph analysis and meta session stitching, capabilities the company says are built to detect session hijacking, token abuse and rogue AI agent behavior in real time.

Why it matters: - WideField Security is targeting a shift in how enterprises defend identities, moving from login-time checks to continuous monitoring of active sessions. - The patents cover technology designed to spot session hijacking, token theft, third-party token abuse and rogue AI agent behavior. - WideField says the approach is meant for an environment where AI agents and other non-human identities operate alongside human users across SaaS, cloud and on-premises systems.

What happened: - WideField Security announced June 2, 2026, that the United States Patent and Trademark Office granted two patents covering its identity threat detection and identity security posture management technology. - U.S. Patent No. 12,592,942, “Session Analysis for Identity Threat Detection and Identity Security Posture Management,” was issued March 31, 2026. - U.S. Patent No. 12,592,943, “Session Analysis for Identity Posture Management and Security,” was also issued March 31, 2026. - The company said the patents protect foundational methods for ingesting telemetry, building live session graphs, stitching meta sessions and enforcing remediation policies in real time.

The details: - The live session graph technology ingests continuous telemetry from identity-adjacent sources including Secure Access Service Edge, Endpoint Detection and Response, Identity and Access Management and cloud applications. - The system builds a graph of nodes and edges representing users, devices, applications, sessions and third-party app tokens. - When analysis surfaces a security event, the platform can revoke sessions and access tokens, disable federated logins and quarantine downstream applications. - The meta session stitching technology is meant to connect fragmented activity across distributed sessions, including an Okta SSO session, downstream Salesforce and GitHub sessions, an OAuth-connected third-party app and the underlying device. - The method correlates those signals into a unified signal hierarchy, reconstructs the full session underlying each meta session and classifies posture as healthy, potentially compromised or compromised. - WideField said the patents cover telemetry across SaaS, cloud, on-premises, IAM, EDR, MDM and SASE sources. - The company said the technology is designed to build stitched meta sessions for every AI agent, non-human identity and human identity.

Between the lines: - WideField is arguing that identity security is becoming a continuous runtime problem rather than a one-time authentication event. - The company pointed to longer-lived application sessions and said a stolen token can now give an attacker prolonged access. - WideField also framed Agentic AI as a new identity category because AI agents and autonomous workflows can hold credentials, chain API calls and run without human supervision. - Kartik Kumar, co-founder and CTO, said the value comes from correlating IAM login events, downstream application activity, EDR malware detections, SASE network anomalies and OAuth grants into one composite view of identity activity.

What’s next: - WideField said the patented methods are part of its platform for real-time identity visibility, threat detection and response. - The company says the platform is intended to keep enforcing security policy as identity risk is detected across active sessions. - Enterprises facing credential abuse and AI-driven access complexity are the clearest target for this approach.

The bottom line: - WideField is trying to own the technical foundation for continuous identity defense as stolen tokens, long-lived sessions and autonomous AI agents reshape enterprise risk.

More information is available at WideField Security and on WideField Security on LinkedIn.