CrowdStrike Launches Continuous Identity Security for AI Agents

CrowdStrike unveiled Continuous Identity for AI Agents on June 15 at Identiverse 2026, expanding its Falcon Next-Gen Identity Security portfolio with a capability designed to continuously authorize AI-driven actions based on identity, context, and risk.

The cybersecurity company said the new offering moves beyond traditional access controls that rely on static policies and standing privileges. Instead, every action performed by an AI agent is evaluated in real time based on who owns the agent, who initiated the request, and the security posture of the associated device.

Elia Zaitsev, Chief Technology Officer at CrowdStrike, said autonomous AI systems require a different approach to identity security than conventional enterprise applications.

“AI agents are transforming how work gets done, and how identities must be secured,” Zaitsev said. “Authorize once and trust indefinitely is not a security model; it's a liability.”

The announcement reflects a broader shift across the cybersecurity industry as organizations deploy AI agents capable of accessing data, invoking tools, interacting with APIs, and coordinating with other agents without direct human intervention.

Recent industry discussions have increasingly focused on the risks associated with non-human identities and autonomous software systems. Growing enterprise investment in AI governance and security controls highlights how organizations are preparing for those challenges as agentic AI adoption expands.

Securing AI Agents With Continuous Authorization

CrowdStrike said Continuous Identity for AI Agents dynamically grants, denies, and revokes access based on real-time risk signals. The capability is powered by technology acquired through the company's purchase of identity security startup SGNL earlier this year.

According to CrowdStrike, each AI agent receives a cryptographically verifiable identity using the SPIFFE (Secure Production Identity Framework for Everyone) standard, which is designed to replace static credentials such as API keys with secure workload identities.

The company said authorization decisions are made using contextual information, including agent ownership, request origin, and device risk posture. When an agent delegates a task to another agent, the system preserves that context throughout the chain.

CrowdStrike also said the platform eliminates standing privileges by granting access only when required and revoking it once the task is completed.

The launch comes as vendors across the identity security market increasingly position AI agents as identities that require governance, lifecycle management, and policy enforcement rather than treating them solely as software tools.

Falcon Platform Expands AI Security Focus

CrowdStrike said the new capability integrates with Falcon AI Detection and Response (AIDR), which continuously analyzes prompts and intent to identify permission abuse, suspicious behavior, or attempts to manipulate large language models beyond their authorized scope.

When risky behavior is detected, the platform can revoke access before actions are executed, according to the company.

The offering extends CrowdStrike's identity security controls across human users, machine identities, and AI agents operating in on-premises, cloud, browser-based, and software-as-a-service environments.

The announcement also represents one of the first major product rollouts tied directly to CrowdStrike's acquisition of SGNL, a deal valued at approximately $740 million. At the time of the acquisition, CrowdStrike said the transaction would strengthen its ability to deliver continuous, risk-aware authorization across all identity types.

As enterprises accelerate AI deployment, security vendors are increasingly focused on ensuring that autonomous systems can be authenticated, monitored, and controlled throughout their lifecycle. CrowdStrike's latest launch underscores how identity security is evolving from one-time access decisions toward continuous verification and authorization for AI-powered operations.