Skip to Content

MC1457685 Task Manager features for visibility into AI workloads

Summary

  • Windows Task Manager now shows deeper visibility into AI workloads, including per-process NPU and GPU neural engine activity on the Processes and Performance tabs.
  • The change is relevant to IT professionals, developers, and power users who need to troubleshoot performance, validate AI model placement, or understand battery/performance behavior.
  • This is a visibility enhancement on Windows devices; it does not indicate a required configuration change or service disruption.
  • Microsoft recommends reviewing the new Task Manager metrics and the linked guidance article to understand how to use the new data.

Microsoft 365 Message Center ID: MC1457685
Primary Service: Windows
Admin Impact: Low
User Impact: Low
Release Start: 19 Aug 2026
Release End: 19 Aug 2026
Last Modified: August 20, 2026
Category: Stay informed
Tags: New feature, Admin impact, Admin action
Status: Launched
Products & Platforms: Windows

History

8/19/2026 Item Added to Message Center

Microsoft Message

You can use Windows Task Manager for deeper visibility into AI workloads running on a device. Specifically, per-process activity for the neural processing unit (NPU) and graphics processing unit (GPU) neural engine now appear on the Processes and Performance tabs, alongside other familiar data.

Rollout schedule

  • These Task Manager enhancements became available on some newer devices in the May and June 2026 non-security updates.
  • A new blog article is now available to illustrate different use cases for IT professionals, developers, and power users.

Impact on your organization

As AI workloads become more common on Windows devices, visibility into NPU and GPU neural engine utilization can help you make better informed decisions. This data can help as follows:

  • For IT professionals: Troubleshoot and report on performance.
  • For developers: Validate whether AI models are landing on the intended NPU or GPU neural engine instead of CPU or standard GPU compute.
  • For power users: Understand and improve performance and battery life.

Action required/recommendations

Explore the new metrics in Task Manager to evaluate application performance, troubleshoot resource bottlenecks, and better understand how your devices use AI-capable hardware. Read Task Manager features for visibility into AI workloads.

Compliance considerations

No compliance considerations are identified. Review as appropriate for your organization.