Nudge Study Finds AI Agents Span Multiple Platforms

Research from Nudge Security reports that AI tools now connect directly to company data in ways traditional security controls often miss.

Updated on Nov. 17, 2026 in Artificial Intelligence

Isometric editorial illustration of a server corridor with cable bundles branching toward an open doorway in deep teal and mustard, evoking AI tools reaching company systems unseen.
Nudge Security's new report finds AI agents are connecting to company data through everyday business platforms, often beyond the reach of conventional identity and security controls. AI Illustration. Upload story photo >

Nudge Security has published AI Adoption in Practice 2026: Beyond the Approved List, a research report based on anonymized, aggregated telemetry from its customer environments. The report found that AI agents are operating inside the business platforms employees use every day.

Why it matters

Because these tools reach company data outside the visibility of conventional identity and security systems, organizations may have limited insight into what is acting on their information. The report indicates the gap between what identity providers can see and what actually accesses company data is expected to keep widening.

The research relies on direct telemetry observation of AI tools, integrations, browser extensions, and agents, with adoption and integration data current as of September 2026 and agent data as of October 2026. More than half of organizations using AI agents run them across multiple platforms, though the report does not detail the full scope of the dataset.

The players

Nudge Security

A security company that published the AI Adoption in Practice 2026 report based on telemetry from its customer environments.

The details

Nudge Security's report, AI Adoption in Practice 2026: Beyond the Approved List, examines how AI tools connect to business systems outside formally approved channels. Researchers based the findings on direct observation of usage across customer environments rather than self-reported surveys.

Timeline

  1. September 2026: Adoption, integration, and browser extension data was collected.

  2. October 2026: Agent data was collected.

  3. November 17, 2026: Nudge Security publicly announced the research report.

The Tech Race

This report extends Nudge Security's first AI adoption study by adding newer customers and tracking how AI agents now operate across platforms. It shows that the trend the company documented earlier has continued into agent-based tools that reach company data beyond traditional security oversight.

Employees using AI tools connected to work accounts may be granting access to company data that IT teams do not monitor. Organizations may need to review which AI agents and browser extensions are operating inside their systems.

The takeaway

Knowing which AI tools are actually running inside your workplace software is now a practical security question. Ask your IT or security team what agents and extensions have access to shared files and accounts.

Further reading

For more on how these tools are reshaping workplace software, see the Artificial Intelligence section.

Source note: This article includes information reported by Star Local.