vmblog logo 2024 wht (updated)

Concentric AI Extends AI Data Security, Introducing Semantic DLP to Desktop AI Applications, Coding Agents, and AI Models

Share: 

David Marshall | Published: November 20, 2026

Concentric AI announced the Concentric AI Semantic DLP Agent, which extends its industry-leading data loss prevention (DLP) capabilities to the endpoint. Concentric AI also introduced the industry’s first conversation intent analysis of AI prompts to detect and protect sensitive data that is inconsistent with a user’s role within the organization.

Concentric AI is the industry’s first company to deliver inline categorization for browser-based or desktop AI to identify and protect sensitive data in a single unified AI and data security governance platform, far more accurately than tools that rely on regex or pattern-based methods. This new functionality not only provides superior data identification and protection but also performs conversation intent analysis to flag AI activity inconsistent with a user’s role within the organization. It supports local AI agents running on the endpoint, such as Copilot, Claude, ChatGPT, Gemini, and Perplexity; coding agents, such as Claude Code, Cursor, Codex, and GitHub Copilot; and other custom AI models running on the endpoint.

“As a public health agency, the protection of sensitive information, particularly protected health information (PHI), is paramount,” said Jason Frame, CIO at Southern Nevada Health District. “Our community entrusts us with some of their most personal information, and maintaining that trust is a responsibility we take seriously. Concentric AI’s Semantic DLP gives us greater visibility into how AI tools are being used across our organization, including shadow AI, while helping us establish guardrails to prevent sensitive information from being exposed through public AI applications. Extending those capabilities directly to the endpoint is another important layer of protection, allowing us to embrace the benefits of AI while safeguarding the information and public trust we are responsible for protecting.”

While DLP has existed for years, most data security vendors recognize the growing need to extend these protections to AI and are developing or acquiring technologies to meet this need. Concentric AI provides organizations with comprehensive visibility and control over data at rest, data in motion, and all AI use, all within a single unified AI and data security governance platform. The ease of full policy deployment, combined with out-of-the-box capabilities, enables customers to gain comprehensive visibility into AI use and protect sensitive data from being shared with AI in as little as four weeks.

“The business wants productivity, and security teams need a way to say yes to AI safely,” said Dhruv Jain, Chief Product Officer at Concentric AI. “Breaches happen at the seams between security products, where two tools describe the same data differently and write policy their own way. Every seam is a vulnerability that attackers exploit today at incredible speed. We are eliminating those seams with a single data taxonomy, common data security intent, and policy framework. The same Concentric AI Semantic Intelligence Engine™ that understands your petabyte-scale data at rest now protects it inline in real time at the moment an employee pushes it to an AI application, or an AI agent reads or uploads it.”

Organizations are eager to use a wide range of AI applications to innovate, improve productivity, and reduce costs. However, security teams have quickly realized they lack visibility into which AI tools are used across their organizations and have limited ability to prevent sensitive data from being shared in prompts. Once confidential data is shared with these models, it is impossible to know when or how it could surface in responses to users outside the company. Many organizations have relied on disparate tools to secure their entire data estate, inevitably creating gaps in visibility and protection where one tool ends and another begins.

Concentric AI’s unified AI and data security governance platform enables organizations to deploy a single, unified platform and policy framework to protect data at rest, data in motion, and data used by AI. This technology performs inline categorization with the Semantic Intelligence Engine, which leverages Concentric AI’s proprietary language models to identify and protect sensitive data far more accurately than tools that rely on regex or pattern-based methods.

Availability

The Concentric AI Semantic DLP Agent is available today from Concentric AI and its authorized partners.