NordLayer Launches NordLayer AI Scout to Expose Shadow AI Across the Workplace

NordLayer has launched NordLayer AI Scout, a new AI visibility and security solution that shows companies which AI tools employees are actually using across web and desktop, including services that were never formally approved.

The product brings sanctioned and unauthorized AI use into one view, tracks how different tools are being used, and can inspect and help control what employees type into AI prompts. It also maps the connections used by AI agents, giving IT teams a clearer picture of what those systems can reach once they are running.

The launch comes as corporate AI spending rises faster than many companies can keep track of day-to-day use. NordLayer cites Writer research showing that 59% of companies now spend more than $1 million a year on AI technology, yet a University of Melbourne report found that 70% of employees who reported using AI for work at one point still use free public AI tools at work. Only 42% who reported using AI use company-provided tools.

That gap is what NordLayer AI Scout wants to expose.

“What companies are facing today is an AI zoo — dozens of unsanctioned LLMs in daily use across the workforce, each one widening the organization’s exposure to data leaks and security breaches,” said Andrius Buinovskis, VP of product strategy at NordLayer.

NordLayer AI Scout monitors activity on-device without requiring employees to manually report which tools they use. Security teams can then see where approved AI has taken hold, where workers have adopted alternatives on their own, and which licenses may be sitting unused.

That last part gives the platform a cost-management role as well. NordLayer says NordLayer AI Scout can measure active versus paid-for licenses and estimate token spend across deployed LLMs, which may reveal a very different picture from the software inventory finance or IT already has.

Security Moves Into the Prompt Box

NordLayer AI Scout has also built data loss prevention around one of the easiest ways sensitive information can leave a company: an employee pasting it directly into an AI chat.

Traditional DLP systems were largely designed around files or information moving through known channels. AI prompts create a different problem because the employee may never upload a document at all.

NordLayer AI Scout can inspect prompts before submission and apply company rules at that point. Depending on policy, the system can warn the user, redact sensitive material, or block the prompt entirely.

“We built NordLayer AI Scout for how AI actually works — not just what it looks like from the outside, but for what’s happening at a technical level,” said Vakaris Noreika, VP of product research and development at NordLayer. “Its AI DLP is designed to inspect prompts and control them pre-submission, to ensure that sensitive data is never inputted in the first place.”

That gives security teams something more precise than a blanket ban on unfamiliar AI services. They can see how a tool is being used and decide where controls belong rather than treating every unsanctioned LLM as equally risky.

NordLayer points to IBM data showing that breaches involving shadow AI add roughly $400,000 to the reported global average breach cost of $4.99 million. The problem becomes harder to contain when new AI products appear faster than companies can review them one by one.

Agents Add Another Layer

NordLayer AI Scout is also designed for the next stage of workplace AI, where software does more than answer questions.

Coding agents can connect to outside tools and internal resources through Model Context Protocol servers. NordLayer AI Scout keeps a live catalog of those MCP connections and an inventory of the skills available to each agent, giving IT teams a way to see what the software can actually access.

That matters because approving an agent does not automatically tell a company what sits behind it. An agent may be connected to data or tools that carry more risk than the interface itself suggests.

NordLayer AI Scout is treating those agent connections as part of the same visibility problem as shadow AI. Knowing which tool exists is useful; knowing what it can reach is far more actionable.

View the NordLayer AI Scout demonstration here.

NordLayer AI Scout is available now to organizations. The launch follows NordLayer’s introduction of its enterprise browser earlier this year and pushes the company further into AI-specific workplace security.

For companies already writing AI policies, the immediate problem may be simpler than the policy itself. Before they can decide what employees should be allowed to use, they need to know what employees are using already.

Digital Trends partners with external contributors. All contributor content is reviewed by the Digital Trends editorial staff.