Anthropic's Claude Enterprise Integrates with Over 100 Security Tools for AI Chat Monitoring
Anthropic's Claude Enterprise now offers a Compliance API, allowing over 100 security vendors to monitor AI chats and uploaded files for sensitive data.
Anthropic has launched a significant enhancement for its Claude Enterprise AI model with the introduction of a Compliance API. This new feature enables over 100 security and compliance vendors to integrate Claude's activity logs directly into their existing monitoring and security tools. This move aims to provide organizations with greater visibility and control over the sensitive data that may be processed or shared within AI-assisted workflows.
The Compliance API provides a comprehensive feed of Claude Enterprise activity, including full conversation transcripts, details of uploaded files, and project information. It extends to Cowork and Claude Code sessions, capturing prompts, responses, and even the specific tool calls made during these interactions. Furthermore, the API logs administrative actions such as user sign-ins, setting changes, and administrator modifications, offering a robust audit trail for security and compliance purposes.
This integration allows security teams to scrutinize what employees are inputting into Claude, including pasted content and data shared in prompts. For developers using Claude for coding, the API provides insights into the tools and functions invoked during their sessions. The company is also piloting session records for its Microsoft 365 add-ins and Claude Science, further expanding the scope of monitorable AI interactions.
Among the notable security vendors integrating with the Compliance API are CrowdStrike, Microsoft Purview, Splunk, Palo Alto Networks, Cloudflare, and Zscaler. These partnerships span a wide range of security disciplines, including data loss prevention (DLP), Security Information and Event Management (SIEM), identity management, eDiscovery, and AI security posture management (ASPM). Some partners, like Hadrius and Shield, focus on archiving conversations for financial firms with strict recordkeeping obligations.
However, the level of data visibility varies by vendor. For instance, Salt Security and Torch Security will not have access to the actual conversation content, while Datadog will ingest only audit logs from the Claude Platform. Enterprise and Platform customers can both leverage the API, but Platform customers receive only activity events, excluding prompts and model responses. On Claude Enterprise, the API can only be enabled by the Primary Owner, with administrators having no access to these settings.
Availability of these integrations is staggered. Netskope's integration is currently in private preview, and Okta's is slated for a beta release to select customers. KnowBe4 anticipates general availability in November 2026, while Vanta's integration is in beta for select customers, with general availability expected in January 2027. This phased rollout allows vendors to refine their integrations and ensure a smooth deployment for their clients.
The introduction of the Compliance API addresses a growing concern among enterprises regarding the security and compliance implications of adopting large language models. As AI tools become more integrated into daily business operations, the ability to monitor and govern their usage is paramount. Anthropic's proactive approach with this API aims to build trust and facilitate the secure adoption of AI across various industries, particularly those with stringent regulatory requirements.