Each year since 2021, we have examined how the U.S. Patent and Trademark Office Patent Trial and Appeal Board (PTAB) has been deciding appeals of § 101 rejections from examiners.[2] And so far, the numbers have been remarkably similar. 2024 was no exception -- applicants continued to face affirmance rates that were not only very high, but stubbornly consistent with prior years. The story here is not one of progress or meaningful change, but of inertia where an entrenched status quo provides predictability by denying patents for a wide range of inventions.
The overall affirmance rate for 2024 was 88.6%. As shown below, this is in line with affirmance rates from previous years.
Put another way, about 7 out of 8 appeals from examiner rejections on grounds of § 101 in 2024 were affirmed by the PTAB. Conversely, the likelihood of a successful PTAB 101 appeal is 11.4%. This is approximately the same chance of flipping a coin 3 times and getting no heads (12.5%), the rate of left-handedness in the U.S. (~10%), or your cat coming when you call it (scientifically unverifiable but feels about right).
The persistently high affirmance rate of § 101 rejections at the PTAB is best understood as a direct consequence of the Supreme Court's decision in Alice Corp. v. CLS Bank. The two-step framework articulated in Alice provides little concrete guidance for practitioners, examiners, or judges. The Court deliberately avoided defining what constitutes an "abstract idea" or when an "inventive concept" suffices to transform such an idea into patent-eligible subject matter. This vagueness has left lower tribunals and the USPTO with a test that is inherently malleable.
The Federal Circuit's subsequent case law has not resolved the ambiguity. Instead, it has produced a patchwork of inconsistent precedents, where eligibility may hinge on subtle distinctions that are difficult to reconcile. Cases such as Enfish v. Microsoft and McRO v. Bandai suggest that claims directed to specific technological improvements can survive § 101 scrutiny, while decisions like Electric Power Group v. Alstom or ChargePoint v. SemaConnect represent how readily claims can be characterized as unpatentably "abstract." The result is a body of law where outcomes depend heavily on framing and analogy, not clear statutory or judicial criteria.
While individual examiners at the USPTO apply § 101 in a notoriously inconsistent and personalized fashion, the PTAB consistently errs on the side of ineligibility. The PTAB's affirmance rate is thus less a reflection of the merits of applicants' arguments and more a predictable byproduct of an eligibility test that lacks objective criteria and is skewed against applicants. The USPTO's recent subject matter eligibility memo merely reiterates the vagueness of the test, failing to provide anything close to a clear distinction between eligible and ineligible inventions.[3]
In any event, our data gathering process was as follows. From the PTAB's search interface, we specified the following criteria: decision dates between January 1, 2024 and December 31, 2024, a proceeding type of "appeal", a decision type of "decision", and an issue type of "101". These search results were overinclusive and thus required filtering and a degree of manual review.
We only included decisions in which the applicant appealed an examiner's Alice-based § 101 rejection and the PTAB ruled on this basis of appeal. To simplify the analysis, we considered a decision to be an "affirmance" of the examiner's § 101 rejection if the PTAB held at least one claim invalid under § 101. Excluded cases where those in which: (i) the appeal was not of a § 101 issue but the PTAB set forth a new grounds of rejection based on § 101, and (ii) the appeal was of a § 101 issue but the PTAB did not decide the case on § 101 grounds (e.g., the examiner withdrew the § 101 rejections after the appeal was filed). This gave us 692 qualifying appeals, of which 613 resulted in affirmance on § 101 grounds.
In addition to the abysmal overall affirmance rate of 88.6%, we broke the data down by USPTO technical center (TC). For sake of comparison, we also included corresponding data from 2023 and 2022 in the table below (click on table to enlarge). To provide a sense of scale, entries in italics are for technical centers with 40 or more 101 appeals for that year.
As expected, TC3600, where most business method inventions submitted to the USPTO land, continues to have the majority of appeals (488 of 692) as well as the worst affirmance rate. This shows that appeals from this TC are for applications that the PTAB treats as virtually per se ineligible. TC3700, which also handles some business methods, is nearly as bad, reinforcing the point that the PTAB views any claim drawn to so-called "non-technical" features as almost certainly doomed. But high affirmance rates are evident across nearly all art units in which software-implemented inventions are examined, and even life sciences inventions are far from safe. The persistence of these numbers year after year confirms that the PTAB's application of § 101 is not a matter of discretion or shifting doctrine, but a structural feature of an unworkable test.
Another set of factors we considered were the grounds under which claims were found to be abstract by the PTAB.[4] In line with USPTO practice, a claim is "abstract" when it recites mathematics, mental steps, or methods of organizing human activity. Of the affirmances, 18.3%, 63.9%, and 73.41% were based on these categories, respectively. Compared to 2023's numbers of 17.3%, 46.7%, and 68.1%, we see mathematics holding steady, while there was a sharp uptick in affirmances based on the mental steps category.
We have noticed, at least anecdotally, that the mental steps category is appearing with greater frequency in rejections of AI-based inventions. One explanation could be that the PTAB is simply reviewing more appeals in this technology area. The methods of organizing human activity category, which is already high, ticked up further. AI-based inventions directed to automation of tasks previously performed by a human may be a cause of this.
Finally, we considered situations where the applicant appeals on grounds other than § 101 (e.g., §§ 102 or 103) but the PTAB decision included a sua sponte § 101 rejection. This occurred 9% of this time in 2024, which is consistent with the 10% of 2023. For applicants, the practical consequence is clear. Even if you overcome the hurdles of §§ 102 and 103 at the PTAB, there is a non-negligible chance that § 101 will be raised against you for the first time. Such a rejection increases the cost, delay, and uncertainty for the application, and reinforces the systemic disadvantage patentees face under the current eligibility regime.
In short, applicants faced the same uphill battle in 2024 that they did in the previous years, and nothing in these statistics suggests that the trend will change absent intervention from Congress or the Supreme Court. As Judge Paul Michel has warned, this kind of unpredictability makes patents unreliable assets of questionable value. The consequences fall heaviest on startups and small businesses, which depend on enforceable patents to raise capital and challenge incumbents. The current affirmance rates all but guarantee the opposite, fortifying the position of dominant players while draining the lifeblood from emerging companies.
[1] Walter DeGroft was a 2025 Summer Associate at MBHB. He is currently a law student at the Chicago-Kent College of Law.
[2] Patent Docs has published our previous articles for 2021, 2022, and 2023.
[3] https://www.uspto.gov/sites/default/files/documents/memo-101-20250804.pdf
[4] The vast majority of the § 101 appeals were of software, business methods, and engineering-related technologies. Life sciences § 101 appeals were mostly based on the "law of nature" exclusion to patentability rather than the "abstract idea" exclusion. Very few (less than 10) life sciences appeals were in this data set.
[View source.]
