Alex Moss | March 17, 2022

Under former Director Andrei Iancu, the Patent Office changed its examination guidelines in ways that made it easier to get patents on subject matter that is not eligible for patent protection—abstract ideas, laws of nature, and natural phenomena—than it had been in years. But apparently, those changes were still not making it easy enough. Now, the Patent Office has begun implementing a program, the “Deferred Eligibility Response Pilot Program,” that reverses the order of patent examination by putting questions of patent eligibility last instead of first. This approach is bad for patent quality and effective patent examination. If the program becomes permanent, the result will be even more new patents that allow private companies to extort small businesses, distort or thwart competition, and impede public access to existing tools and knowledge.

Patent eligibility limitations are crucial safeguards for ensuring that the patent system promotes more innovation than it deters. By prohibiting patents on subject matter that humans could not have invented, these limits help prevent patents from blocking public access to knowledge that is, and should remain, in the public domain. These changed rules inPatent Office’s analysis of patent eligibility have already led examiners to grant patents on ineligible subject matter. For example, the Patent Office granted a patent on the basic idea of using networked computers to communicate health information. Allowing companies to monopolize ideas that are uninventive yet essential—especially for people seeking access to health care in rural areas or in the midst of a pandemic—endangers public health without encouraging or rewarding technological advancement. 

By implementing this pilot program, the Patent Office is changing examination procedures in ways that will magnify the public danger patents present. The Deferred Eligibility Response Pilot Program lets examiners ignore questions of patent eligibility until after they have considered all other issues. Although the pilot program is already ongoing, it is not yet permanent, and the Patent Office has requested public comments before taking that step. Thanks to Harvard Law School’s Cyberlaw Clinic, PIPLI and the ACLU submitted comments explaining why the pilot program is flawed and should be paused for meaningful review before becoming permanent. The comments were authored by students Cosmas Robless (JD ‘23) and Mariah Bellamoroso (JD ‘23), working closely with Clinical Instructor Kendra Albert.

Our comments explain that the pilot program, adopted at the behest of Senators Thom Tillis and Tom Cotton, is misguided at its core. In requesting a program such as this, the senators asserted that patent eligibility questions are hard for examiners to decide consistently and predictably. The data shows the opposite: patent examiners are more likely to be affirmed on patent eligibility questions than other kinds of decisions on appeal (like obviousness and written description). That shows they are deciding these issues consistently, predictability, and that their rejections on eligibility grounds are overwhelmingly correct. There is no rationale for making examiners save these questions for last. 

 Nor are there any practical benefits. In fact, there are substantial downsides. Patent eligibility issues are less time-consuming for examiners to decide than other issues. Deciding patent eligibility requires the examiner to apply the law to a patent application without researching or considering other technical references.  By contrast, deciding issues like obviousness require searching for analogous technical references and then comparing them, element-by-element, to each and every claim in a patent application. Given that examiners are already under extremely tight time constraints, with approximately 19 hours to review an application, they should be encouraged—not discouraged-- from deciding the least-time consuming issue first. 

Determining eligibility issues last also tips the scales in favor of finding the application eligible. That’s because it requires the examiner to assume eligibility throughout the rest of their examination. On a practical level, an examiner must conceptualize a claimed invention and compare it to one or more technical references. By taking these steps, an examiner effectively assumes there is a patent-eligible invention that can be used as the basis of such comparisons. 

It is extremely unlikely that an examiner who has identified such an advance on the assumption of patent eligibility will be able to consider the question of patent eligibility as fairly and open-mindedly as an examiner evaluating that question first. Studies from clinical and social psychology and behavioral economicsdemonstrate that initial assumptions have a strong influence on subsequent decision making: the influence of a previous judgement, even implicit, can be observed even when new information is provided that directly contradicts the previous assumption. 

By making patent examination less efficient and less impartial, the Patent Office’s experimental approach creates a substantial risk that more ineligible patents will be issued. Those ineligible patents can then be used to demand license payments and initiate litigation that small businesses, nonprofits, and individual developers cannot afford. The result is less innovation, less competition, higher barriers to entry, and fewer consumer options at higher prices. There is no reason to risk these harms to the public, especially as our country is in the midst of attempting to recover from the COVID-19 pandemic, reduce inflation, and establish peace in Europe without sacrificing a nation’s existence or starting a world war. 

We hope the Patent Office will pause the pilot program after its initial term, carefully examine the resulting data, make that data available, and consult with the public again before making this problematic change permanent. If it results in more low quality patents being issued, as seems likely, the greatest costs will be borne, as ever, by the public.