Exabeam doubles AI detection coverage and adds Anthropic Claude support

2026-07-01 · Source: AI – SiliconANGLE · Field: Technology & Digital — Cybersecurity & Data Privacy, Artificial Intelligence & Machine Learning · Depth: Intermediate, quick

Summary

Exabeam Inc. expanded its security operations platform on July 01 2026, significantly enhancing its capabilities for detecting and investigating artificial intelligence agents. The update doubles AI-focused detection coverage to 90 and adds monitoring support for Anthropic PBC's Claude, alongside existing coverage for OpenAI Group PBC's ChatGPT, Google LLC's Gemini, and Microsoft Corp.'s Copilot. New detections target anomalous interactions, unauthorized autonomous activity, suspicious prompt behavior, and denial-of-wallet indicators, also covering shadow AI and unauthorized configuration changes. The release includes coverage aligned with the OWASP Top 10 for Agentic AI within Outcomes Navigator and extends Nova AI with a Rules Creator for natural language rule building and Nova Related Cases for linking shared entities. Additionally, Exabeam released Observra, an open-source telemetry layer for AI agents, which complements Praxen, an open-source agent verification tool released on June 23. The platform also expands LogRhythm SIEM integrations and adds phishing email ingest.

Key takeaway

For MLOps Engineers or AI Security Engineers deploying autonomous AI agents, your security posture must account for agent-specific behaviors. You should integrate solutions that provide deep visibility into AI agent interactions, tool invocations, and consumption patterns to detect anomalous activity that traditional security tools miss. Consider utilizing open-source tools like Observra for runtime telemetry and Praxen for pre-deployment verification to ensure comprehensive agent security and compliance.

Key insights

Security platforms must evolve to monitor autonomous AI agent behavior, which often mimics legitimate user activity.

Principles

• AI agent activity requires continuous behavioral analysis.
• Pre-deployment verification complements runtime observability.
• Security tools should align with established frameworks like OWASP.

Method

Observra captures AI agent activity across frameworks, normalizes it into events, enriches with cost/risk signals, and routes to security platforms.

In practice

• Use Nova Rules Creator to build detection rules in natural language.
• Employ Praxen for pre-deployment AI agent verification.
• Utilize Observra for comprehensive AI agent telemetry.

Topics

• AI Agent Security
• Behavioral Analytics
• Anthropic Claude
• OWASP Top 10
• Security Operations Platform
• Open-Source Security Tools

Best for: CTO, VP of Engineering/Data, AI Architect, AI Security Engineer, MLOps Engineer, Director of AI/ML

Related on AIssential

• 🗞️ Anthropic launched Claude Code Security to scan your code repositories for bugs and suggest security patches. — AI is advancing in code security, robotics platforms, and developer tools, yet faces challenges in user adoption and intellectual property protection.
• How we contain Claude across products — Robust, multi-layered sandboxing is crucial for containing AI agent actions and preventing data exfiltration.
• Claude Enterprise Meets the Security Graph: Wiz Integrates with Anthropic's Compliance API — Integrating AI tool activity logs into existing security graphs centralizes visibility and governance.
• Anthropic Just Leaked Their Entire Codebase (By Accident) — A major code leak from Anthropic's Claude Code revealed advanced agentic AI engineering patterns and unreleased features.
• Anthropic is launching a new AI model for cybersecurity - The Verge — Anthropic's Claude Mythos Preview autonomously identifies and exploits system vulnerabilities, enhancing cybersecurity defenses.
• Honeycomb introduces agent observability features to keep an eye on production — Observability for AI agents in production is crucial for understanding their behavior and impact.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by AI – SiliconANGLE.