Anthropic Turns Claude Into a First-Class Citizen in the Security Stack
Anthropic is moving Claude firmly into the center of enterprise security operations with 28 new integrations built on the Claude Compliance API. As AI tools become part of everyday workflows, organizations need to govern them with the same rigor as any other critical application. The Compliance API exposes two key data streams: Claude Enterprise conversation content—such as chats, uploaded files, and projects—and activity events from both Claude Enterprise and the Claude Platform, including logins, admin changes, and configuration updates. Security teams can plug this data into existing DLP, SIEM, SASE, data security, and observability tools to enable continuous monitoring and automated policy enforcement instead of manual exports and periodic audits. New integrations with vendors such as SailPoint, Varonis, and others allow IT to apply established security policies, audit practices, and regulatory controls directly to AI usage, reducing blind spots and accelerating enterprise AI governance.
Varonis Extends Data-Centric AI Security and Oversight for Claude
Varonis is using the Claude Compliance API to bring Claude Enterprise and Claude Platform activity into its Atlas AI Security Platform, extending its data-centric approach into AI workflows. As AI agents access data at machine speed, security teams need visibility into what those systems are doing and which data they can reach. Through the integration, Atlas ingests Claude Enterprise conversational content and administrative events, allowing teams to monitor chats, file uploads, and projects, detect anomalies in real time, and maintain detailed audit records. For the Claude Platform, Atlas tracks how assistants and agents are built and operated, storing audit events, flagging risky behavior, and stress-testing agents for issues like prompt injection. Crucially, Atlas connects this AI activity to underlying data sensitivity and access entitlements, turning Claude usage into something that can be governed with the same policies and controls enterprises already rely on for broader data security.
SailPoint Brings Identity Security Integrations to the AI Workforce
SailPoint’s new Claude Compliance API connector pushes identity security directly into the AI layer, treating Claude Enterprise like any other critical application in the identity perimeter. By integrating Claude into the SailPoint Identity Security Cloud, organizations gain unified visibility over Claude users, groups, roles, and now non-human identities such as AI agents. This lets teams apply consistent governance policies, certify access, and reduce the risk of Shadow AI—unguarded AI usage that sits outside official controls. The connector enables enterprises to discover and register Claude agents as part of a single agent registry, then apply adaptive, context-aware access decisions based on who or what is accessing which resources, when, and why. For security leaders, this means AI platforms no longer sit in an identity blind spot. Instead, AI access and entitlements can be governed with the same enterprise-grade identity security models used for human users and other machine identities.
From Compliance Gap to Continuous Enterprise AI Governance
Together, the Claude Compliance API and its growing ecosystem of identity security integrations and data security tools are turning AI governance from a manual, ad hoc exercise into a continuous, automated discipline. By exposing both conversation content and detailed activity events, the API lets organizations embed Claude into their existing security controls, rather than building an entirely new stack. DLP platforms can inspect prompts and outputs, SIEM and observability tools can correlate Claude events with other signals, while identity platforms like SailPoint enforce least-privilege access for both humans and AI agents. Vendors such as Varonis add data context, mapping AI activity to sensitive information and access rights. This integration-first approach directly addresses enterprise concerns around AI adoption: auditability of AI decisions, regulatory reporting, enforcement of internal policies, and provable controls over who and what can use powerful AI systems at scale.