Regarding USPTO Proposal 2: Automated Pre-Examination Search

Regarding USPTO Proposal 2: Automated Pre-Examination Search 1
Adam J. Gianola 2
Kilpatrick Townsend & Stockton LLP

We generally support the U.S. Patent and Trademark Office (USPTO) having a modern automated search tool to place the most relevant prior art before examiners at an early stage of examination. Such a tool may increase examination quality and resultant patent quality while decreasing overall pendency of applications.

However, such a tool may have downsides. If examiners rely only on the search tool’s results and do not perform full examinations, examination quality may decline, potentially increasing overall pendency. Additionally, examiners might accord excess weight to the search tool’s results regardless of their actual relevance simply because the results were uncovered by the automated tool, which again could decrease overall examination quality.

Training programs introducing examiners to the automated search tool must address these risks. The USPTO should adopt guidelines defining the role of the automated search tool and its results, and instructing examiners how to properly use the references generated. These guidelines should be publicly available so applicants can understand how examiners use, combine, and weigh the results of the search tool.

Additionally, details on the how the search tool operates should be publicly disclosed. For example, the USPTO should clearly identify which search algorithms are available or utilized in analyzing and searching applications, how the contents of applications are extracted and used, and how keywords and concepts in an application are identified. This information will enable applicants and the public to understand the tool’s abilities and how the USPTO uses it.

In addition, whenever the tool is used, the application’s file wrapper should document the specific algorithms utilized, the parts of the application extracted, the keywords and concepts identified, and the results generated by the tool. This will clarify the record in a way similar to the current practice of listing examiners’ search queries in appendices to Office Actions.

To further enhance quality, we propose several additional modifications to the Enhanced Patent Quality Initiative Automated Pre-Examination Search proposal.

Proposal: Allow Public Access to Any Newly Utilized Automated Search Tool
The public should have access to any new automated search tool utilized by the USPTO. Public access will allow applicants to quickly and efficiently conduct prefiling searches, providing them with the ability to identify any novelty-destroying prior art and enabling them to draft original claims with a focus and scope appropriate in view of the prior art. If applicants use the search tool prior to filing to revise, refine, or otherwise improve their inventions and the associated disclosures in their patent applications, the quality of new applications will improve and prosecution time may decrease.

Further, if details on the search tool’s use are laid forth in the file wrapper as described above, applicants will be able to replicate the automated search if needed.

Proposal: Allow Applicants to Amend Claims After the Automated Search is Performed and Prior to Examination
Since the results of the automated search will be generated prior to examination, we propose immediately providing the applicant with the results of the automated search tool and allowing the applicant to make a preliminary amendment before examination proceeds. For example, the USPTO could create a two month window after the automated search results are sent to the applicant, during which the applicant may make preliminary amendments in light of the search results. Such a procedure will enhance the benefits of the automated search tool by expediting prosecution.

Proposal: Periodic Evaluation of Automated Tool and Publication of Related Quality Metrics
Finally, the USPTO should periodically or continuously evaluate the performance of the automated tool to ensure that it is operating effectively. This may include requesting feedback from examiners, applicants, and practitioners, or comparing manual search results with the tool’s results. For example, a quality committee could review an examiner’s manual search results for a case and compare them to the automated search results to determine the accuracy and thoroughness of the automated tool. As another example, the committee could periodically run the automated tool on an expired patent or an abandoned application and compare the automated results to the results from the examiner’s search of record. The USPTO should publish whatever metrics it generates on the quality of the automated tool.

  1. A prior version of this comment was submitted to the U.S. Patent and Trademark Office on May 6, 2015, in response to its Request for Comments on Enhancing Patent Quality, 80 Fed. Reg. 6475 (Feb. 5, 2015).
  2. This comment is also supported by Kate S. Gaudry, Angel Lezak, Matthew T. Kitces, Richard B. Almon, and Thomas D. Franklin. It is attributable only to the author and indicated supporters, and does not represent the opinions or beliefs of any other individuals, companies, or organizations.