USPTO Adopts Compass AI to Speed Patent Reviews

The agency is deploying new AI agents to help reduce the patent review process from five years to one year.

Updated on Oct. 20, 2026 in Artificial Intelligence

Isometric editorial illustration of a dense, uniform stack of document folders representing institutional patent reviews.
The U.S. Patent and Trademark Office has deployed a new AI tool, Compass, to automate routine patent checks and reduce years-long backlogs. AI Illustration. Upload story photo >

Live Poll

Do you trust government agencies to utilize artificial intelligence to manage their internal workloads?

The U.S. Patent and Trademark Office has implemented an agentic AI tool named Compass to handle routine formality checks on patent submissions. This rollout aims to streamline operations and address a massive backlog of 700,000 pending applications.

Why it matters

By automating rote tasks, the agency intends to free up its 9,000 patent examiners to focus on complex analysis and human-led interaction. The initiative also seeks to standardize review quality across staff with varying levels of experience.

The Compass tool is designed to perform formality checks on patent submissions, while the previously implemented Scout chatbot assists with research. These agents support a workforce of approximately 9,000 examiners.

The players

U.S. Patent and Trademark Office

This federal agency is responsible for granting U.S. patents and registering trademarks.

Barry Schindler

He serves as the Acting Commissioner of the U.S. Patent and Trademark Office.

The details

The agency is integrating AI to manage technical bottlenecks and improve consistency in review quality among personnel, which includes experts like PhD biochemists. Compass operates as an agentic tool that shoulders repetitive duties to shorten the current five-year review period.

Timeline

  1. October 20, 2026: Acting Commissioner Barry Schindler discussed the agency initiative at a summit.

The Tech Race

The adoption of Compass reflects a broader shift toward integrating autonomous agents into complex bureaucratic workflows. This move positions the agency to replace legacy manual review processes with AI-driven systems to maintain competitiveness in global patent processing.

Applicants can expect a more efficient interaction process as examiners shift their focus toward substantive human communication. These technological improvements may ultimately decrease the wait time for individuals and companies seeking patent protections.

The takeaway

The move demonstrates how government agencies are increasingly relying on specialized AI agents to manage massive data backlogs. This transition highlights a growing trend of prioritizing automation to scale human expertise in highly technical professional fields.

Further reading

For additional context, visit the Artificial Intelligence section.

Source note: This article includes information reported by Nextgov.

Live Poll

Do you trust government agencies to utilize artificial intelligence to manage their internal workloads?