Posted in | News | Artificial Intelligence

Kovrr Launches AI Interaction Data Fabric to Secure Enterprise AI Workflows

Add AZoRobotics on Googleas a preferred source

KovrrSep 11 2026Reviewed

Kovrr, a leading provider of AI and cyber risk, security, and governance solutions, announced the AI Interaction Data Fabric, the correlation layer that powers its AI Security and Governance Platform. The fabric began with network telemetry, where the strain of AI traffic hit first and hardest, and has grown into a connected set of sources spanning browser, endpoint, agent, cloud, LLM, identity, and DLP.

AI agents are becoming a standard feature of the software that enterprises already run. In fact, Gartner predicts that 40% of enterprise applications will feature task-specific AI agents by the end of 2026, up from less than 5% in 2025. As employees reach for shadow AI tools and enterprises put agents to work across their workflows, sensitive customer records and regulated financial data are routinely transmitted into external systems outside the organization's control.

The wider web shows the same pattern. In June 2026, Cloudflare recorded automated requests crossing 57.4% of worldwide HTTP traffic, with human users falling to 42.6%. Across the same period, HUMAN Security measured agentic AI traffic growing 7,851% year over year.

With Kovrr's new AI data fabric, while each and every source reports accurately on the AI activity it can see, the correlation layer assembles all of it into one account, complete enough to build a continuously updated inventory of AI in use, monitor how those systems behave, enforce policy on their use, and price the exposure they create.

"Security teams have no shortage of AI telemetry at this point. What they do lack is a way to discern what all of it implies for their risk posture," said Yakir Golan, CEO and Co-founder of Kovrr. "One console records that an interaction occurred, another holds the identity, and a third separately holds the content. Each reports an action that appears permitted, although none can establish the significance of that activity taken together."

Every AI Interaction. Full Context. One Platform.

Kovrr's AI Interaction Data Fabric applies correlation rules that identify which AI signals belong to the same interaction, so activity recorded seconds apart in separate consoles is reconstructed as one sequence. Each sequence then establishes facts no source holds alone, such as the identity behind an interaction, the category of data it moved, or the vendor receiving it.

The same sequence details how far the exposure reached, whether that covers every grant a vendor holds across the estate or the exact set of labeled documents an assistant opened. A response can then meet the full extent of what happened instead of the portion one console happened to record.

The Evidence, Published in the Open

Highlighting the risks that stay invisible without triangulation, Kovrr publishes the AI Interaction Data Fabric Insights, a series documenting one cross-source detection at a time. The first issue follows a finance employee who pasted classified financial data into a consumer AI account, an action four separate tools each recorded as ordinary. Connected in the AI data fabric, however, the same signals established a regulated-data exposure and stopped it before the content was submitted.

"Illuminating these multi-layered exposures is the first step toward understanding how this new kind of risk behaves," said Golan. "Kovrr's AI Security and Governance Platform exists to give enterprises that essential insight, drawn from the systems they already run, ensuring potentially material risks are accounted for and mitigated."

Source:

Kovrr