Concentric links compliance to Claude

Concentric AI said on May 15 it had tied its semantic data-loss-prevention software to Anthropic’s Compliance API for Claude, giving customers a way to inspect prompts, responses and policy violations as they move through the model. The San Jose, California, company described the release as an Anthropic integration that lets enterprises audit Claude activity directly through a compliance interface rather than periodic exports. Anthropic has separately said its Compliance API gives organizations programmatic access to usage data and customer content for observability, auditing and governance. Together, the two announcements show how AI security vendors are starting to build controls around the model layer itself rather than only around files, endpoints or network traffic. ### What exactly did Concentric say it can see inside Claude? Concentric AI said the integration lets organizations use its Semantic DLP product to audit “prompts, responses, and policy violations” across Claude platform activity. The company’s product page also says customers can gain visibility into prompts, responses and attachments flowing through Claude Enterprise in browser and app environments. (concentric.ai) The May 15 announcement said the system is meant to strengthen data security governance for companies using Claude. Concentric framed the product around identifying sensitive data exposure and enforcing policies tied to generative AI use, rather than only scanning stored data after the fact. ### Where does Anthropic’s Compliance API fit in? Anthropic said when it introduced the Compliance API for business plans that enterprise organizations could use it to meet regulatory requirements with “real-time programmatic access” to Claude usage data and customer content. (concentric.ai) The company said that access was designed to support continuous monitoring and automated policy enforcement systems. Anthropic’s Trust Center says the company positions its enterprise controls around secure practices and transparency. (concentric.ai) Its public developer materials also show a broader push to make Claude easier to deploy in production environments, including admin controls and business-plan tooling. ### Why are prompts and responses the focus instead of just files? Concentric AI said the integration audits the actual exchanges with Claude, not only the documents that may have been uploaded before or after a model interaction. (anthropic.com) That matters because sensitive information can appear in a prompt, be transformed in a response, or be exposed through attachments and generated outputs during use of the model. That framing comes from Concentric’s own product description and announcement. (trust.anthropic.com) A Concentric guide published earlier this year said employees often upload contracts, financial records and regulated data into Claude, and argued that exposure can begin with the surrounding data environment before a model processes the material. That document is marketing material, but it shows the company’s stated rationale for connecting DLP controls to Claude activity itself. ### Is the “industry-first” claim independently verified? (concentric.ai) Concentric AI called the release an “industry-first Anthropic Compliance API integration” in its press release. Reuters could not independently verify whether no other vendor had previously integrated with Anthropic’s Compliance API in the same way, and Anthropic’s public materials reviewed here do not make that specific ranking claim. (concentric.ai) What is verifiable from public sources is narrower: Anthropic has a Compliance API for enterprise customers, and Concentric says it has built a product integration on top of it for Claude governance. Other security vendors have also publicly discussed Anthropic’s Compliance API since its release, though those references do not establish who was first. ### What does this add to Concentric’s broader product lineup? (concentric.ai) Concentric AI’s website says the company sells a data security governance platform focused on autonomous data discovery, classification, risk monitoring and remediation across data at rest, data in motion and enterprise generative AI. The Claude integration extends that positioning into Anthropic’s ecosystem with a named control point for AI interactions. An August 2025 press release from Concentric described earlier “industry-first” integrations with Wiz, Salesforce and GitHub. (anthropic.com) The Claude announcement fits that pattern of using integrations to widen the company’s reach inside enterprise security workflows. ### What happens next for customers evaluating the product? Concentric AI said the Claude integration is part of its generative AI security offering and has posted product and demo information on its website. (concentric.ai) Anthropic’s business-plan materials say the Compliance API is available for organizations that need programmatic monitoring and governance controls. Companies evaluating the setup now have named product pages from both vendors and a May 15 launch date for the Concentric release. (concentric.ai 1) (concentric.ai 2)