The United States Patent and Trademark Office (USPTO) has announced it will begin an AI-powered search pilot program for utility patent applications. According to a draft Federal Register Notice, the agency will start accepting petitions to join the program on October 20, 2025, to assess how sharing automated search results affects the early stages of patent examination.
Key Takeaways
- The USPTO is launching the "Automated Search Pilot Program" to test an AI search tool.
- The program will provide applicants with a list of potential prior art before the official examination begins.
- Petitions will be accepted from October 20, 2025, through April 20, 2026, or until capacity is met.
- The pilot is limited to 1,600 applications, with a cap of 200 for each technology center.
- The goal is to evaluate if early AI-driven insights improve the efficiency of the patent application process.
USPTO Announces Automated Search Pilot Program
The U.S. Patent and Trademark Office is set to introduce a new initiative called the Automated Search Pilot Program. The program's primary function is to evaluate the effects of providing patent applicants with AI-generated search results before a formal examination of their application is conducted.
This initiative represents a significant step in integrating artificial intelligence into the patent review system. The USPTO aims to understand whether early access to these automated results can help applicants make more informed decisions about their filings.
According to the official notice, the feedback and data gathered during the pilot will help the agency determine the effectiveness of using AI tools to streamline the initial phases of patentability assessment.
How the AI Search Tool Functions
Once an application is accepted into the pilot, the USPTO will utilize a specialized AI tool to conduct a preliminary search for prior art. This tool analyzes several key components of the patent application to find relevant documents.
The AI considers the application's classification under the Cooperative Patent Classification (CPC) system, along with the full text of the specification, claims, and abstract. It uses this information as context to search through extensive public databases.
What is Prior Art?
In patent law, "prior art" refers to any evidence that an invention is already known. It includes patents, publications, or any other public disclosure that shows the invention existed before the filing date of a patent application. Identifying relevant prior art is a crucial step in determining if an invention is new and non-obvious enough to be patented.
The databases searched include U.S. Patents, U.S. Pre-Grant Publications (PG-Pubs), and the Foreign Image and Text (FIT) collection, which contains publications from various international patent authorities. The AI then ranks the findings from most to least relevant.
Applicants will receive an Automated Search Results Notice (ASRN) identifying up to 10 of the most relevant documents found by the tool. The USPTO has clarified that this notice is not an official action under 35 U.S.C. 132, and applicants are not required to respond to it.
The results of the automated search may inform a decision on how to proceed with the application.
This information could lead an applicant to file a preliminary amendment, defer the examination, or even withdraw the application by filing a petition for express abandonment.
Program Eligibility and Application Process
Participation in the Automated Search Pilot Program is subject to specific criteria. Only certain types of applications are eligible for consideration.
Eligibility Requirements
- The application must be an original, non-continuing, nonprovisional utility application filed under 35 U.S.C. 111(a).
- Applications must be filed on or after October 20, 2025, and on or before April 20, 2026.
- Continuing applications, such as divisional or continuation-in-part applications, are not eligible.
- Submissions must be made electronically through the USPTO’s Patent Center.
- Applicants are required to be enrolled in the Patent Center Electronic Office (e-Office) program.
Program Capacity Limits
The pilot program is capped at a total of 1,600 applications. This total is distributed across all of the USPTO's Technology Centers (TCs) that handle utility applications. Each TC is slated to accept up to 200 applications. The USPTO noted that the pilot could be terminated early if any single TC significantly exceeds its 200-application limit.
The petition window will remain open until April 20, 2026, or until the capacity limit for each technology center is reached, whichever occurs first. This structured approach allows the agency to gather a balanced dataset across different technological fields.
Goals and Impact on the Patent Examination Process
The central goal of the pilot program is to collect data on the practical impact of AI-assisted search results on the patent prosecution lifecycle. The USPTO wants to measure whether providing these results at the outset leads to a more efficient and focused examination process.
By giving applicants a preview of potential obstacles, the agency hopes they can address issues earlier, potentially reducing the number of office actions and overall examination time. This proactive approach could save time and resources for both the applicants and the USPTO examiners.
Throughout the pilot, the USPTO will actively solicit feedback from participants. This input will be crucial for refining the AI tools and determining the future role of automated searches in standard examination procedures. The program serves as a real-world test for integrating advanced technology into the agency's core operations.
AI Integration as a Priority for USPTO Leadership
This pilot program aligns with the stated priorities of the current USPTO leadership. Recently confirmed Director Kathi Vidal and her administration have emphasized the importance of leveraging artificial intelligence to enhance operational efficiency and improve the quality of patent examinations.
The Automated Search Pilot Program is a tangible example of this strategy in action. It reflects a broader trend across government agencies to explore and adopt AI technologies to modernize processes, manage workloads, and improve service delivery. The outcomes of this pilot will likely influence future technology investments and policy decisions at the USPTO.