Digital workforce security pioneer KnowBe4 has announced the expansion of its Agent Risk Manager solution to support Anthropic’s Claude enterprise environment. Unveiled at the Black Hat USA conference, the integration extends KnowBe4’s agent governance architecture to Claude workflows, building on its native capabilities for Microsoft Copilot to deliver real-time monitoring and automated threat detection for autonomous AI agents.
As corporate operations transition from basic prompt-driven tools to autonomous AI agents, security operations teams face expanding visibility gaps. Employees frequently deploy Claude-powered agents to automate complex multi-step tasks, analyze proprietary data, and connect with third-party software-often running outside traditional corporate security frameworks and identity controls.
Addressing the Agentic AI Risk Exposure
The integration addresses a growing governance challenge highlighted in recent industry research. According to KnowBe4’s 2026 Agentic Risk to Human Wins Report, 58% of cybersecurity executives confirm that AI agents actively perform actions within internal business workflows. However, administrative oversight continues to lag behind adoption, with 52% of respondents reporting that AI usage within their organization remains unapproved or ungoverned.
Agent Risk Manager closes this operational blind spot by delivering an easily integrated security layer that monitors and governs agent behavior without altering the underlying Large Language Model (LLM). The system extends KnowBe4’s 15-year risk-based methodology-originally developed for human security awareness—to the emerging non-human workforce.
Also Read: Energage and Engagedly Merge to Deliver Workplace Culture and Talent Management Platform
Key Capabilities for Security Operations Teams
KnowBe4’s Agent Risk Manager for Claude delivers specialized features engineered for modern Security Operations Center (SOC) environments:
-
Real-Time Threat Interception: Utilizes six specialized detection engines to actively analyze agent interactions for prompt injection attempts, PII and sensitive data exposure, unauthorized privilege escalations, and unapproved tool usage.
-
Blast Radius Visualization: Interactive Tool Network mapping renders connected APIs and credentials in a visual node structure, enabling security analysts to identify high-risk access points if an agent becomes compromised.
-
Guided Enterprise Onboarding: Allows security teams to integrate Claude environments quickly through a streamlined onboarding workflow that operates without altering underlying AI model performance.
“The industry spent decades securing the human element and are challenged to protect the newest members of the digital workforce, AI agents,” said Matt Duren, SVP of AI and data at KnowBe4. “It is incredibly important to protect what these agents are doing, especially when employees might be using features like skip permissions mode, which allows the LLM to make its own decisions autonomously with zero oversight. With KnowBe4’s Agent Risk Manager for Claude, we focus on the actions, outputs, and tool usage of these agents, ensuring they do not become an unmonitored backdoor into sensitive company assets.”
