One of the Early Appeal Decisions in Quantum Computing Represents a Positive Datapoint

By Andrew Velzen --

As reported by Quantum Insider[1], this past week, the Patent Trial and Appeal Board (PTAB) at the United States Patent and Trademark Office (USPTO) overturned an examiner's rejections of an application directed to a quantum computing invention.[2]  Even though this decision is based solely on the view of three administrative patent judges at the PTAB, it is notable since there have not yet been a substantial amount of PTAB or Federal Circuit decisions in the technical area of quantum computers.[3]  Further, as multiple commentators have noted, the current landscape under 35 U.S.C. § 101 could potentially be hostile to quantum computing innovations.[4]  Hence, opinions rendered in this space will likely be an important early barometer regarding the likelihood of success in terms of patentability.  Given this, I feel it is worth gleaning what we can from this early decision. During examination, the Examiner rejected the majority of the pending claims under 35 U.S.C. § 112(a) as lacking written description and all pending claims under 35 U.S.C. § 101 as directed to patent-ineligible subject matter. The 35 U.S.C. § 112(a) rejections warrant little discussion.  Essentially, the Examiner had rejected the "generating an objective function" step of the claims (step (a) from claim 1 above) as only demonstrating that the inventors were in possession of the specific objective function described as an example in the Specification.  According to the Examiner, this was insufficient to meet the written description requirement.  The PTAB roundly disagreed with the Examiner's analysis on this point.  In the panel's view, the Examiner:  conflated the enablement requirement with the written description requirement as it applied to using example species to describe a genus; improperly stated that the Specification fails to indicate how the inventors intended for the features to be performed while simultaneously citing to specific examples from the Specification of how to perform the claim features; and cited to irrelevant caselaw in the rejection. Perhaps most importantly with respect to the § 112(a) rejection, the PTAB noted that "the original claim 1, which is a part of the Specification, recites the disputed limitation."  As such, the PTAB overturned the § 112(a) rejections.

Much more interesting, in my view, is the PTAB's addressing of the Examiner's rejections under 35 U.S.C. § 101.  Under the traditional two-step Mayo/Alice test, the Examiner had previously stated that the preparation and measurement of quantum states is mere "gathering data of a particular type or source to be used in performing the abstract idea" and "is an attempt to limit the abstract idea to a particular field of use or technological environment."[8]  In its own analysis, the PTAB jumped directly to Step 2A, Prong 2 of the Mayo/Alice test (i.e., whether the claimed abstract idea is integrated into a practical application).  In doing so, the PTAB determined that the applicant's claims did integrate the abstract idea (e.g., the mathematical relationships) into a practical application and, therefore, are patent-eligible.

In analyzing the applicant's claims, the PTAB agreed with the applicant's argument that the claims provide a technological improvement by enabling noisy quantum computers (which have limited circuit depth) to practically solve linear systems.  In order to arrive at this determination, the PTAB pointed to discussions from the applicant's specification of limitations in other quantum computing systems:

One problem to which quantum computers have been applied is solving linear systems.  Existing techniques for using quantum computers to solve linear systems, however, cannot be implemented on current quantum computers, which are noisy and have low circuit depths.

...

What is needed, therefore, are improvements to quantum computers for solving linear systems.

...

A hybrid quantum-classical (HQC) computer system, which includes both a classical computer and a quantum computer, solves linear systems.

As a result of the above, the Examiner was reversed on all counts.

There are two points I take away from this decision.  The first point is something that most of us who routinely work in technologies likely to face § 101 rejections are already very well-aware of.  That is, make sure to explicitly recite both an underlying technical problem and how your invention provides a technical improvement when drafting a patent specification.  That way, your first go-to in responding to a § 101 rejection can always be to parrot passages from your own specification to easily argue that your claims provide the requisite technical improvement under Step 2A, Prong Two (or under the "significantly more" analysis of Step 2B) of the Mayo/Alice test.  It is clear that these types of passages and the associated argumentation from applicant's counsel were sufficiently persuasive for the PTAB.

Second, and more crucially / technology-specific in my view, this case is also an example that enabling quantum computers to perform techniques that classical computers already routinely perform can still constitute a technical improvement under § 101.  For example, classical computers can already solve linear systems of equations.  However, here, the goal of the applicant was to enable noisy quantum systems to be used in solving linear systems of equations.  Thus, by framing the application appropriately, even merely enabling some classical computing actions in a quantum computing space can provide the requisite technical improvement.  And why shouldn't it?  Ultimately, like here, by accounting for some of the shortcomings of quantum computers (e.g., noise), the speedup of quantum computers can be actualized, which is, in itself, absolutely an improvement.  While it may seem clear that this should be considered an improvement, it is relieving (especially in an area of the law rife with self-contradictions and head-scratchers) that this first result agrees.

Obviously, though the claims at issue here do involve mathematics at a baseline level, the claims still recite actively manipulating qubits in the all-important "controlling" step that the PTAB highlighted.  It will be intriguing to see, as we go forward, whether claims that are one or more steps further abstracted (e.g., relating more tenuously to hardware and more directly to quantum algorithms) will also be viewed favorably under § 101.  Also, as I noted upfront, this is just one case with three APJs, so we will have to monitor if this decision reflects the general opinion toward quantum innovations going forward or instead marks an aberration.  I remain hopeful that, like the invention in question, judges and examiners evaluating quantum innovations can continue to block out the (§ 101) noise.

[1] https://thequantuminsider.com/2025/02/17/patent-board-overturns-rejection-of-hybrid-quantum-computing-method/

[2] The full text of the PTAB decision can be found here.

[3] And what prior jurisprudence does exist is relatively uninstructive / inapplicable in a general sense.  For example, in the nonprecedential Federal Circuit decision in In re Huping Hu (Fed. Cir. 2021), four patent applications relating to alleged quantum entanglement technologies were rejected primarily under 35 U.S.C. §§ 101 and 112(a) as lacking operability, written description, and enablement.  These final rejections were upheld by the PTAB and then upheld by the Federal Circuit.  In that case, the examiners, the PTAB, and the Federal Circuit were all incredulous based on the applicant's insufficient description of the inventions.  "The PTO, as the nation's guardian of technologic invention, must be receptive to unusual concepts, for the core of invention is unobviousness.  However, concepts that strain scientific principles are properly held to a heightened standard, typically measured by reproducibility of results.  Here the Board was presented with an apparent departure from conventional scientific understanding, and the Board appropriately sustained the examiners' requirements for experimental verification.  The Board applied a reasonable and objective standard, and acted reasonably in sustaining the examiners' requirements.  Should further investigation."  In re Huping Hu (Fed. Cir. 2021).

[4] See, e.g., my 2023 article from Inside Quantum Technology News – https://www.insidequantumtechnology.com/news-archive/how-to-develop-a-strong-quantum-patent-portfolio-part-1/

[5] https://www.hpcwire.com/2024/10/14/zapata-computing-early-quantum-ai-software-specialist-ceases-operations/ 

[6] Published Specification (U.S. Pat. App. Pub. No. 2020/0104740) at Paragraph [0046].

[7] See also Published Specification (U.S. Pat. App. Pub. No. 2020/0104740), Paragraphs [0050]-[0053].

[8] PTAB Decision, Page 12.

Comments

[Patent Practitioner]

Good article and an encouraging result. In my experience, the PTAB seems to be more receptive to this argument (i.e., technical improvement described in the specification and achieved by the claims under Step 2A Prong 2) than most examiners, which unfortunately adds time and cost to the prosecution.

[kotodama]

Nice article and helpful comment as well. I just note that some of the article content seems to be formatted as images instead of text for some reason? For example, the paragraph immediately following Claim 1 that starts: "As detailed in the specification, ..."

[Don]

The paragraphs presented as images contained special characters that did not display properly as text.

[kotodama]

Thanks, Don, that makes sense. It does have the mild side effect of disabling the links to the footnotes that were in the first such affected paragraph. Also the text is smaller and harder to read than the rest of the article. I guess in an ideal world only the the special characters themselves would be rendered as images, but maybe such a feature isn't available in this blog publishing platform.