The U.S. Patent and Trademark Office (USPTO) organizes its examining corps into technical centers (TCs).  Each TC is dedicated to one or more general technical fields.  In some cases, one TC may include two or more unrelated fields, while some fields (software, for example) are represented in multiple TCs.  Currently there are nine TCs, not counting one dedicated to reissues.

Each TC is further subdivided into art units (usually 4-10) and examiners are assigned to these art units.  Thus, examiners tend to focus their efforts on examining patent applications within a fairly narrow field.

Over the nine years since 2014's Alice Corp. v. CLS Bank Int'l Supreme Court case, the number of 35 U.S.C. § 101 rejections has dramatically increased throughout the USPTO.  Anecdotally, however, many practitioners believe that certain TCs are much more likely to result in 101 rejections than others.  In practice, it is generally accepted that TC 3600 -- the TC to where most business-related inventions are routed -- issues the most Alice-based 101 rejections.

We tested this hypothesis by gathering data on Alice rejections per TC from 2020.  We had to look back this far due to the USPTO's 18-month publication delay causing not all relevant data from 2021 to be available.  Here, an Alice rejection is a rejection of at least one claim on the grounds of the claim(s) not meeting the judicial patent eligibility requirements of Alice.  This is a slightly narrower scope than § 101 rejections in general, which include claims rejected on grounds of failing to meet the statutory requirements and grounds of double patenting.

While the raw numbers of Alice rejections per TC in 2020 were not hard to obtain, they were also not very insightful when considered in isolation.  To be able to compare Alice rates across TCs, we also collected a count of § 103 rejections per TC in 2020.  By considering the ratio of § 103 rejections to Alice rejections, we are able to obtain a clearer understanding of how "difficult" each TC is with respect to § 101.  Particularly, using § 103 rejections (which are quite common) as a baseline, we can get a clear understanding of how many Alice rejections to expect per § 103 rejection.

As shown above, the ratio of § 103 to Alice rejections varies wildly across the TCs.  For instance, in TC 1600, you can expect one Alice rejection for about every fifteen § 103 rejections.  On the other hand, in TC 3600, you can expect one Alice rejection for about every three § 103 rejections.  Indeed, TC 3600 has the highest likelihood of an Alice rejection by far when compared to § 103 rejection volume.  This appears to confirm our anecdotal observations.

Other TCs that have relatively high rates of Alice rejections are 1600 and 2100.  Both of these TCs receive a high volume of software applications.  In contrast, TCs 1700, 2600, and 2800 have the lowest rates, with 1700 being quite low.

Based on experience, we expected a higher rate of Alice rejections in TC 3700.  In particular, gaming inventions get routed to this TC, and we have seen frequent Alice rejections for their applications.  This TC, however, also examines mechanical and medical inventions, which typically are not subject to § 101 challenges as often.  Thus, across TC 3700 as a whole, Alice rates fall about in the middle of the field.  This suggests that it might be informative to further break this data down by art unit in a follow-on exercise.

A word of caution -- while this data suggests that TC is highly influential on whether an application is going to receive a § 101 rejection, there are other factors.  By their very nature, some inventions (e.g., business methods, data processing) are more prone to a § 101 rejection than others (e.g., mechanical devices, chemical compounds).  Inventions with a high likelihood of triggering an Alice analysis may receive a § 101 rejection regardless of its TC.

Still, if a goal is to avoid TCs with higher Alice rates, this data provides a compelling view of just how much each TC relies on Alice rejections when compared to § 103 rejections.  As such, TC prediction and steering tools can be advantageous when used in conjunction with the numbers above.