Anthropic has launched Inference Hooks in beta for Claude Enterprise, allowing enterprises to integrate their own security vendors to enforce data policies before prompts reach the model. This shift changes how companies evaluate and adopt AI models, focusing more on control than inherent safety.
TL;DR
- Anthropic's Inference Hooks enable enterprises to integrate their own security vendors to enforce data policies before prompts reach Claude.
- This development shifts the focus from model safety to the level of control security teams have, potentially reducing AI security spending.
- The feature is currently in beta and limited to prompt-side inspection, with six security vendors integrated at launch.
What happened
On August 5, 2026, Anthropic launched Inference Hooks in beta for Claude Enterprise. This feature allows organizations to route every governed prompt through an AI security server before the model processes the request. If the security server denies the request, it never reaches the model.
The integration happens at the infrastructure level, providing coverage across the entire Claude ecosystem, including claude.ai, Claude Cowork, Claude Code, and connected tools across web, desktop, and CLI channels. Requests are signed with an organization's secret, and any denials are logged directly in the Activity Feed.
Six security vendors were integrated at launch to support this capability: Check Point (Workforce AI Security), Cyera (Agent Guardian), Akto, Reco, Proofpoint (DLP), and Metomic.
Why it matters
This development changes the calculus of AI adoption for procurement and security leaders. Previously, the primary question was which model is safe enough. Now, the focus is shifting toward which model gives security teams the most control.
By integrating directly with existing security stacks, organizations can consolidate their AI security spending and potentially reduce the costs associated with maintaining separate proxy or agent-based monitoring tools.
This is a departure from the earlier Compliance API, which focused on auditing activity after the fact. Inference Hooks are designed to stop policy violations before they occur.
Key facts
- Inference Hooks are currently in beta for Claude Enterprise.
- The feature allows enterprises to integrate their own security vendors to enforce data policies before prompts reach Claude.
- Six security vendors were integrated at launch: Check Point, Cyera, Akto, Reco, Proofpoint, and Metomic.
- The integration happens at the infrastructure level, providing coverage across the entire Claude ecosystem.
- Requests are signed with an organization's secret, and any denials are logged directly in the Activity Feed.
- The current protocol scope is restricted to prompt-side inspection only.
- The system is limited to an allow-or-deny verdict and does not currently support prompt rewriting or redaction.
- The feature is exclusive to Claude Enterprise and is not available on Amazon Bedrock or Google Cloud.
Context
Security teams are currently caught in a difficult balancing act. They are being pushed to adopt agentic AI at breakneck speed, yet the tools to keep that technology on a short leash are lagging behind. According to Deloitte, 74% of organizations plan to adopt agentic AI within two years, but only 21% have established mature governance models to manage the risks.
Gartner projects that 40% of enterprise applications will embed AI agents by the end of 2026. This rapid expansion is already showing cracks, with industry reports indicating that AI-related security incidents have climbed 55% year-over-year, and 35% of organizations admit they would be unable to shut down a rogue AI agent if one were to appear.
