Claude Enterprise gives security teams visibility: a security center compares settings to Anthropic's recommendations, and inference hooks check prompts before they run
Synopsis
Anthropic added a security center, inference hooks, centralized management and new access controls to Claude Enterprise, so security teams can see how their organization's configuration differs from Anthropic's recommended setup, send prompts and tool calls to existing data security tools for a check before they run, and manage identity and access from one place. The security center and centralized management remain in beta while inference hooks are generally available; the security center is guidance based on recommendations, not a certification or a guarantee of compliance.
Interpretation
Claude Enterprise gains a set of admin features aimed at security teams, covering configuration review, checks at inference time, cross-organization identity management and member access control. Previously security teams had to inspect Claude's configuration setting by setting, and lacked a single entry point for applying existing data security policies to prompts and tool calls before they execute. The security center gathers security settings for chat, Cowork and Claude Code on one page, showing for each control the current value, the recommended value where one exists and a link to change it, and it rechecks the configuration each time the page loads.
Inference hooks connect Claude to the data security tools an organization already uses, so data loss prevention policies apply to prompts and tool calls. Prompt hooks send prompts from supported Claude products to the security tool for a check, while pre-tool-call hooks check tool calls before they run. Both kinds of hooks are generally available and work with security vendors, so teams keep the policies they have already written.
Perspective
The features target Claude Enterprise security staff and admins, some are in beta with exceptions listed in the Help Center, and the security center is guidance based on Anthropic's recommended configuration, not a certification or a guarantee of compliance with any law or standard.
The security center and centralized management are in beta, as are signing members out through the Admin API, the self-serve IP allowlist and Claude Desktop settings management; the EDR integration is in beta on macOS and Windows, sensors run in detection mode and report activity without blocking it, and it covers the virtual machine on the user's computer, not Cowork sessions in the cloud.