Exaforce Expands AI Security Tool to Monitor Multiple Platforms Beyond Claude

Exaforce is addressing the evolving needs of enterprise security teams with its latest tool, designed to discover and monitor AI agents without requiring additional endpoint sensors. By leveraging telemetry data that organizations already collect, the company aims to provide a more integrated approach to AI security.

The Exaforce AI Security tool combines insights from various agentic AI platforms with data from endpoints, cloud services, SaaS applications, and code repositories. This comprehensive data integration allows the system to identify potential risks, detect unusual behavior, and respond effectively to threats. Co-founder Ariful Huq emphasized that the tool utilizes data already collected by Security Operations Centers (SOCs) to audit every AI application and associated agent. It connects each to the personnel, devices, and permissions involved, enabling the detection of misuse and threats at a granular level.

Expanding on its Claude Compliance API integration announced earlier this year, Exaforce now includes monitoring capabilities for additional AI models such as those from OpenAI, Gemini, and Microsoft Copilot. This extended functionality allows organizations to correlate AI agent activities with existing SOC data, gaining insights into what each agent is doing, who is operating them, and whether their actions may pose security risks.

Osterman Research’s Principal Analyst Michael Sampson highlighted that monitoring AI agents is crucial since these agents interact with various devices, data sources, and identities. He argued that existing solutions like Endpoint Detection and Response (EDR), Identity Access Management (IAM), and model-provider logging are often insufficient in isolation. "Something needs to unify behaviors and actions across the board to assess what's appropriate and what isn't," he noted.

Importantly, Exaforce emphasizes that no new gateways, browser extensions, or endpoint agents are necessary for data collation. Instead, it gathers information from existing EDR systems, usage logs from model providers, and productivity suite activities, constructing a contextual understanding of AI behavior.

Beyond Passive Monitoring

Exaforce’s capabilities go beyond simple observation. Once a potential threat is identified, the tool can employ existing controls within EDR systems and model-provider administration to take immediate action. This includes revoking sessions, deactivating model-provider API keys, isolating devices, or terminating processes associated with flagged agents.

While competitors are opting for distinct strategies, Exaforce’s integrated approach positions it uniquely in the security landscape. For instance, Palo Alto Networks’ Prisma AIRS 3.0 emphasizes centralized visibility and policy enforcement specific to the agent lifecycle. Similarly, SentinelOne has launched its Prompt AI Agent Security to uncover risks and enforce least privilege security policies. Meanwhile, CrowdStrike’s Falcon Guardian, unveiled in September, seeks to enhance endpoint detection for AI threats.

Nevertheless, recent studies indicate that merely identifying AI agents is insufficient for effective security. A survey conducted by the Cloud Security Alliance in March 2026 revealed that 68% of organizations struggle to differentiate between human-operated activities and those performed by AI agents. Additionally, 74% of respondents noted that their AI agents often had unnecessary access, with 52% acknowledging instances where these agents inherited permissions initially granted to humans.

This complexity highlights the inadequacy of existing solutions which primarily focus on observation and posture management, lacking robust in-line blocking or remediation capabilities. Independent analyst Avivah Litan pointed out that effective security will necessitate discovering both sanctioned and unsanctioned agents across multiple clouds, mapping human and machine ownership, and enforcing policies even when agents operate outside their originating platforms.

Exaforce's Huq mentioned that the AI Security tool is evolving towards achieving these goals. By integrating data from various AI and agent sources into a framework that provides necessary context, the tool aims to accurately differentiate between human identities and those of AI agents that may have assumed them.

The Exaforce AI Security platform is available now, either for self-operation or through the company’s Managed Detection and Response (MDR) service, offering enterprises the opportunity to enhance their AI monitoring capabilities significantly.