“Streamlining” Examination—New USPTO Pilot Programs

United States Patent and Trademark Office (USPTO) Director John Squires has stated that “it is to the benefit of all [patent system] stakeholders [that] prior art is identified and applied at the earliest stage of examination … as we benefit as a society from patents ‘born strong,’ beginning with the original patent grant.”

On Oct. 27, 2025, the Federal Register published an announcement that the USPTO under Squires is implementing a new limited pilot program: the Streamlined Claim Set Pilot (SCS Pilot) Program. This follows another pilot program announcement published in the Federal Register on Oct. 8, 2025, the Artificial Intelligence Search Automated Pilot (ASAP) Program.

SCS Pilot Program

The SCS Pilot Program looks to “trade extra claims for expedited feedback” by setting a requirement that a qualifying original nonprovisional application include only a single independent claim, with up to nine dependent claims. An already-filed noncontinuing application can be amended by preliminary amendment to meet this requirement.

An applicant must file a petition to be considered for the SCS Pilot and pay a nominal $150 fee for large entities. If approved to enter the pilot, an application will advance out of turn onto a special docket until a first office action is issued. Thereafter, the application will be placed on the assigned examiner’s regular docket.

The program is set to expire on Oct. 27, 2026, or until each of the eight utility Technology Centers have docketed at least 200 applications under the SCS Pilot, a threshold likely to be satisfied very quickly, as expedited prosecution programs (even those with significantly higher fees) have historically been popular. For example, earlier this year in response to applicant demand, the USPTO increased the number of Track One petitions accepted per fiscal year from 15,000 to 20,000. The SCS Pilot, by contrast, is slated to accept only 1,600 petitions.

ASAP Program

The ASAP Program will leverage an internal artificial intelligence (AI) search tool to automatically conduct a preliminary prior art search. The AI search is intended to provide applicants and examiners with a first-pass search that can inform whether a preliminary amendment should be filed or whether an application should be abandoned in order to recover certain filing fees.

Petitioners accepted into the program will receive an Automated Search Results Notice (ASRN) of the 10 “most relevant” results as evaluated by the internal AI tool. A copy of the ASRN will be placed in the application file. The ASRN is not considered a notice or rejection under 35 U.S.C. § 132 and no applicant response is required. Also, references cited in the ASRN are not automatically made of record on a PTO-892 form and are not required to be considered by an examiner. To ensure consideration of these references by the examiner—especially in continuing applications—an applicant must file an Information Disclosure Statement in compliance with 37 CFR 1.97 and 1.98.

Participants in the ASAP program should be aware that, unlike examiner-conducted prior art searches, automated searches under the ASAP program are not primarily focused to the broadest reasonable interpretation of pending claims. Instead, the AI models supporting the tool are trained using “publicly available patent data” and will leverage “the classification of the application under the Cooperative Patent Classification (CPC) system, as well as the specification, including the claims and abstract, of the application as contextual information” to search across U.S. Patents, Pre-grant Publications and the Foreign Image and Text databases. The AI search tools are not configured to search for non-patent literature in any language.

Reliance on initial CPC classification and specification content—rather than primarily relying on claim language—may raise doubts about the likely relevance of references cited in the ASRN specifically because initial classification itself is an automated process that has consistently exhibited significant error. Roughly one in three applications classified by external classification vendors are reclassified, and thereafter often re-docketed, in response to a request of an examiner or supervisor according to an inspector general report from Aug. 30, 2023. There is, as of this writing, no readily accessible data from the USPTO purporting to assess how many further applications receive an inaccurate or imprecise initial CPC classification but are not reclassified before substantive examination begins.

More simply, errors in automated classification may well compound to reduce the usefulness and applicability of references cited in the ASRN.

It is not clear whether an automated search will be refreshed or performed again if an application is reclassified by an examiner once docketed.

As of this writing, the USPTO has not provided model architecture information, nor any information in respect of whether the system is configured for continuous training, periodic retraining, or if the models invoked during the ASAP Program are fixed and, if so, what knowledge cutoff dates may be relevant.

This document is intended to provide you with general information regarding streamlining pilot programs at the USPTO. The contents of this document are not intended to provide specific legal advice. If you have any questions about the contents of this document or if you need legal advice as to an issue, please contact the attorneys listed or your regular Brownstein Hyatt Farber Schreck, LLP attorney. This communication may be considered advertising in some jurisdictions. The information in this article is accurate as of the publication date. Because the law in this area is changing rapidly, and insights are not automatically updated, continued accuracy cannot be guaranteed.