Enablement Challenges in Quantum Computing Patents: Lessons from Recent PTAB Cases

The quantum computing revolution is generating groundbreaking innovations – and equally complex patent prosecution challenges. Among these challenges, enablement requirements under 35 U.S.C. § 112(a) have emerged as a particularly thorny issue for patent practitioners working in this field.

Recent decisions from the Patent Trial and Appeal Board (PTAB) reveal a troubling pattern: even well-represented corporate applicants are struggling to satisfy enablement requirements for quantum computing inventions. Understanding why these applications failed – and what succeeded – offers critical guidance for practitioners navigating this emerging technology landscape.

The Enablement Problem in Quantum Computing

Enablement requires that “the specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains…to make and use the same.” 35 U.S.C. § 112(a).

For quantum computing patents, this requirement poses unique difficulties. The technology is nascent, complex, and rapidly evolving. Examiners and judges may lack deep technical familiarity with quantum mechanics, quantum information theory, or the specific implementations of quantum processors. This knowledge gap creates friction when evaluating whether a specification enables the full scope of broad claims.

As the Federal Circuit has explained, “to be enabling, the specification of a patent must teach those skilled in the art how to make and use the full scope of the claimed invention without ‘undue experimentation’.” Genentech, Inc. v. Novo Nordisk A/S, 108 F.3d 1361, 1365 (Fed. Cir. 1997). The key question becomes: what constitutes “undue experimentation” in a field where even basic implementations may require sophisticated equipment, specialized expertise, and significant development work?

Case Study 1: Ex Parte Nayfeh – When Two Examples Aren’t Enough

Ex Parte Nayfeh (Appeal 2023-000069; Application 16/513,387) illustrates the enablement challenge in stark terms. The invention related to a “multifunctional quantum node device” comprising three main components:

• A semiconductor vacancy qubit structure
• A superconductor quantum memory nanowire coupled with a spin state of the semiconductor vacancy qubit structure
• A superconductor qubit logic circuit

Independent claim 1 recited:

A multifunctional quantum node device, comprising:

a semiconductor vacancy qubit structure;

a superconductor quantum memory nanowire coupled with a spin state of the semiconductor vacancy qubit structure; and

a superconductor qubit logic circuit coupled with the superconductor quantum memory nanowire and the semiconductor vacancy qubit structure, whereby the device is a hybrid device operable as an interface for computing or quantum-entangled networking.

The Examiner rejected the claims for lack of enablement, and the Board affirmed. The central issue was the term “semiconductor vacancy qubit structure.”

The Board’s Analysis

The Board affirmed the examiner’s position on enablement: namely, that while the specification enables “nitrogen vacancy centers in SiC or diamond,” it “does not reasonably provide enablement for any semiconductor vacancy in any material or for any vacancy in any semiconductor material that forms a qubit.” The specification disclosed only two specific examples – nitrogen vacancy centers in diamond and nitrogen vacancy centers in silicon carbide – but the claims encompassed far more.

According to the Board, the claims covered “any semiconductor vacancy in any material or for any vacancy in any semiconductor material that forms a qubit.” This breadth created an enablement problem because the specification did not teach how to implement the full scope of semiconductor vacancy qubit structures without undue experimentation.

The Board conducted a detailed In re Wands analysis, examining factors including:

• Quantity of experimentation necessary: Substantial experimentation would be required to implement vacancy-based qubits in materials beyond the two described examples
• Amount of direction or guidance: The specification provided detailed guidance only for the two specific implementations
• Breadth of claims: The claims were significantly broader than the enabled embodiments
• Predictability of the art: The field was unpredictable, with different materials exhibiting vastly different quantum properties

Appellant argued that providing “two specific examples” with “broad terminology” should provide enablement commensurate in scope with the claims. The Board disagreed, noting that there was no meaningful dispute that the full scope of the claims was broader than what the specification taught.

Critical Takeaway

The Nayfeh decision reveals that in quantum computing cases, describing even multiple specific implementations may not enable broad genus claims. The specification must either:

1. Describe a sufficient number and variety of species to enable the full genus, or
2. Provide general principles or guidance that would enable a person of ordinary skill to implement other species without undue experimentation, or
3. Narrow the claims to the enabled embodiments.

Notably, while the enablement rejection was affirmed, the Board reversed some of the indefiniteness rejections. While the Board found the term “semiconductor vacancy qubit structure” to be indefinite, it found that other terms such as “nitrogen-vacancy center” would be understood by a person of ordinary skill in the art. This provides some comfort: quantum computing terminology, even when technical and specialized, can satisfy definiteness requirements if it has recognized meaning in the field.

Case Study 2: Ex Parte Baughman – When “Using Quantum Entanglement” Isn’t Enough

Ex Parte Baughman (Appeal 2023-002850; Application 16/204,784) presents another instructive failure. The invention involved a deep learning method where the final step synchronized layers of neural network models “using quantum entanglement.”

Independent claim 1 recited:

A method comprising:

selecting, by a computing device, layers from a plurality of external deep learning models;

concatenating, by the computing device, the selected layers from the plurality of external deep learning models to form a core deep learning model;

training, by the computing device, the core deep learning model; and

synchronizing, by the computing device, layers in the core deep learning model with the layers from the plurality of external deep learning models using quantum entanglement.

The Examiner rejected under § 112(a) for lack of enablement and insufficient written description. The Board affirmed both rejections.

The Enablement Problem

The Board found that the specification did not adequately enable the step of “synchronizing…using quantum entanglement.” While the specification disclosed the concept of using quantum entanglement to transfer weights between model layers, it did not provide sufficient technical detail about how to actually implement this quantum synchronization.

The Board explained: “Within the disclosure, there is no description of any specific starting material or of any conditions under which quantum entanglement of deep learning layers can be carried out.”

The Board conducted a Wands analysis (In re Wands, 858 F.2d 731 (Fed. Cir. 1988)), and found several factors weighed against enablement:

• Direction or guidance: The specification provided minimal guidance on how to achieve quantum entanglement between the layers
• Working examples: No working examples demonstrated the quantum entanglement synchronization
• State of the prior art: Evidence showed that quantum entanglement implementations were non-trivial and limited
• Predictability: The art was unpredictable, with significant technical barriers to implementing quantum entanglement

Citing evidence about the state of quantum entanglement technology, the Board noted that even achieving 18-qubit entanglement was a notable research achievement as of the filing date, and that entanglement lifetimes were limited. The Board questioned “how one of ordinary skill in the art can implement quantum entanglement for potentially large sets of data describing layers between the external deep learning models and the core deep learning model as quantum entanglement itself appears non-trivial.”

Appellant argued that paragraph 77 of the specification “spells out to those skilled in the art that the server uses entanglement and the stored weights to keep the layers of the core deep learning model synchronized.” The Board was not persuaded, finding this disclosure insufficient.

The Written Description Problem

The Board also found insufficient written description, explaining that the specification “merely identifies the result of synchronizing using quantum entanglement but…does not demonstrate how to achieve that result.”

This reflects a key principle: describing a desired result is not the same as describing an invention. As the Federal Circuit has stated, “the written description requirement is not met if the specification merely describes a ‘desired result.’” Vasudevan Software, Inc. v. MicroStrategy, Inc., 782 F.3d 671, 682 (Fed. Cir. 2015).

The Board emphasized: “The more telling question is whether the specification shows possession by the inventor of how [the claimed function] is achieved.” The specification’s statement that synchronization could be performed “using quantum entanglement” – but without explaining the mechanism itself – proved fatal.

The Indefiniteness Issue

Interestingly, independent claim 27 of the patent in issue recited synchronizing “using entanglement” without the qualifier “quantum.” The Board found this indefinite because “entanglement” could refer to non-quantum types of entanglement. Claim 1, which included the qualifier “quantum”, was not the subject of an indefiniteness objection.

This demonstrates that quantum computing terminology, when used precisely, can satisfy definiteness requirements – but precision alone does not satisfy enablement.

Critical Takeaway

Baughman illustrates a fatal drafting error: invoking quantum computing concepts as a “black box” solution without explaining implementation details. Phrases like “using quantum entanglement” or “performed on a quantum computer” may sound impressive, but they do not enable an invention unless the specification explains how the quantum computing features are actually implemented and integrated into the claimed system.

The Contrasting Success: Ex Parte Cao

Not all quantum computing applications fail enablement or written description challenges. Ex Parte Cao (Appeal 2024-002159; Application 16/591,239) provides a successful counterexample.

The invention related to a hybrid quantum-classical computer system for solving linear systems of equations. Independent claim 1 recited:

1. A method for preparing a quantum state that approximates a solution x to a linear system of equations Ax=b for a matrix A and a vector b, comprising:

(a) on a classical computer, generating an objective function that depends on:

(1) at least one expectation-value term derivable from the matrix A, and

(2) at least one overlap term derivable from the vector b and the matrix A; and

(b) training a set of circuit parameters θ, comprising:

(1) on a quantum computer, controlling a plurality of qubits, according to the set of circuit parameters θ, to prepare a quantum state |ψ(θ)⟩;

(2) on the quantum computer, obtaining a measured sample…

(3) on the classical computer, generating an estimate of the objective function based on the measured sample; and

(4) on the classical computer, updating the circuit parameters θ, based on the estimate of the objective function, to optimize a subsequent estimate of the objective function.

The Examiner rejected the claim under both § 101 and § 112(a), the latter as a written description rejection. The Board reversed both rejections.

Why Cao Succeeded on 112

The Board found that the specification reasonably conveyed to skilled artisans that as of the filing date, Appellant possessed the claimed invention. Critically, the Board noted:

1. The original claims recited the disputed limitations, demonstrating possession
2. Paragraphs 22, 49, and 57 of the specification described the limitations and provided specific implementation examples
3. The specification “describe[s] an invention understandable to th[e] skilled artisan and show[s] that the inventor actually invented the invention claimed”

The Examiner had argued that the specification described only “two species of this genus” (specific objective functions at paragraphs 49 and 57) and failed to enable the full scope. The Board rejected this analysis, finding the disclosure adequate.

The Key Difference

What distinguished Cao from Nayfeh and Baughman? Several factors:

1. Mathematical specificity: The specification provided specific mathematical formulations for the objective functions and explained how they were derived and applied.

2. Implementation detail: The specification explained the interaction between the classical and quantum computing components, describing the specific steps performed on each platform.

3. Technical explanation: The specification explained why the approach worked – how using quantum computing enabled solving linear systems that classical computers struggled with due to noisy quantum hardware with limited circuit depth.

4. Examples connected to principles: Rather than just listing isolated examples, the specification connected specific implementations to general principles that would enable variations.

The Board noted that the Examiner’s rejection appeared to conflate written description and enablement issues. Although the Board reversed only on written description grounds, the Federal Circuit has recognized that ‘written description and enablement often rise and fall together,’ (Ariad, 598 F.3d at 1352), suggesting the Cao disclosure would likely fare well under an enablement analysis as well.

Critical Takeaway

Cao demonstrates that mathematical algorithms implemented on quantum computers can satisfy enablement and/or written description requirements when the specification provides:

• Specific mathematical formulations
• Clear delineation of which steps occur on classical vs. quantum hardware
• Technical explanation of how and why the quantum computing aspects contribute to solving the problem
• Sufficient detail to connect specific examples to broader principles

Ex Parte Hastings: The “After-Arising Technology” Exception

Ex Parte Hastings (Appeal 2023-003447; Application 16/286,337) presents a fascinating wrinkle in quantum computing enablement analysis.

The invention related to a method of operating a quantum computing device, specifically involving phase estimation techniques on qubits. The Examiner rejected under § 112(a), arguing that the claims were broad enough to encompass an “after-arising technology” – specifically, a large-scale fault-tolerant quantum computer that did not exist and was “more than a decade away.”

The Board reversed, citing the principle that “application sufficiency under § 112, first paragraph, must be judged as of the filing date.” The Board held:

We agree with Appellant. To now say that Appellant should have disclosed the fault-tolerant quantum computer, which on this record did not exist as of the filing date, and whose future existence is purely speculative, “would be to impose an impossible burden on inventors and thus on the patent system.” In re Hogan, 559 F.2d 595, 606 (CCPA 1977).

The Hogan Principle

The Hastings decision relies on In re Hogan, which established that enablement must be judged based on the state of the art as of the filing date. Later-developed technologies cannot be used as a basis for finding lack of enablement.

The CCPA explained in Hogan: “If later states of the art could be employed as a basis for rejection under 35 U.S.C. § 112, the opportunity for obtaining a basic patent upon early disclosure of pioneer inventions would be abolished.”

The Hastings-Baughman Paradox

Hastings and Baughman sit uneasily together, and practitioners should approach the issue of enablement with some caution in light of the tension between this pair of decisions.

On the surface, the two decisions appear to reach opposite conclusions on structurally similar questions. In Hastings, the Board accepted claims directed to phase estimation methods performed on a fault-tolerant quantum computer that admittedly did not exist and was “more than a decade away.” In Baughman, the Board rejected claims reciting “quantum entanglement” as a synchronization mechanism, despite quantum entanglement being a physical phenomenon that has been experimentally demonstrated for decades and is, in practical terms, at least as accessible today as fault-tolerant phase estimation.

The decisions can potentially be reconciled by distinguishing how each claim invoked quantum computing. The Hastings claims recited a specific sequence of quantum operations – phase estimation, state evolution, a second phase estimation, and error evaluation – framed as a method to be performed on a quantum computing device. The Board reversed the enablement rejection on a narrow ground: the Examiner had faulted the specification for not enabling large-scale fault-tolerant quantum computers, but under Hogan, enablement must be judged as of the filing date, and the non-existence of future hardware cannot defeat present enablement of a method. The Baughman claims, by contrast, recited “synchronizing … using quantum entanglement” without specifying how entanglement would accomplish the synchronization: no protocol for encoding classical weights into quantum states, no mechanism for establishing entanglement between separated systems, no measurement scheme for translating entangled states into weight updates. The Hogan principle was not available to rescue Baughman because the problem was not that future hardware was needed to practice a disclosed method – the problem was that the specification allegedly disclosed no method at all, only a desired result paired with an invocation of a physical phenomenon. Hogan protects inventors from being required to disclose later-arising technology; it does not excuse an applicant from disclosing how the claimed invention works using technology available at the filing date.

But how persuasive is that reading? Entanglement is not inherently a “phenomenon” rather than a “method.” A specification reciting specific gate sequences to produce Bell pairs and specific measurement protocols to transfer classical information would recite a “method” just as much as phase estimation does. Conversely, phase estimation at useful scale requires fault-tolerant quantum computing that does not yet exist – a concern the Baughman Board treated as weighing heavily in its Wands analysis when evaluating entanglement-based claims, citing evidence about current qubit counts and entanglement lifetimes. One could reasonably ask why state-of-the-art concerns about entanglement scalability counted against Baughman while state-of-the-art concerns about fault-tolerant hardware did not count against Hastings. The answer may lie in how each Board characterized the gap between disclosure and claim scope: Hastings framed the gap as one of future hardware implementing a disclosed method (squarely within Hogan’s protection), while Baughman framed the gap as one of missing disclosure about how to practice the claimed synchronization at all (outside Hogan’s reach). Whether that characterization is doctrinally stable, or merely reflects the panels’ differing intuitions about how “algorithmic” each specification read, remains an open question.

The honest answer is that the two decisions may represent genuine uncertainty in how the PTAB will treat quantum computing inventions at different levels of abstraction. The outcomes may turn as much on how “algorithmic” versus “phenomenological” a specification reads to the deciding panel as on any bright-line doctrinal rule. A specification written in recognizable computer-science terms – named algorithms, specific operations, gate-level protocols – appears more likely to benefit from Hogan’s protection against after-arising hardware critiques. A specification that gestures at quantum phenomena functionally, without grounding the gesture in specific protocols, appears more likely to face Wands-factor scrutiny that effectively imports state-of-the-art concerns. (How these two approaches would fare under 101 analysis is left as a possibly disheartening exercise to the reader.)

If practitioners can take any practical guidance from these two decisions, it may be to keep in mind the following two points. First, draft defensively, extensively, and in specific detail: recite specific, named quantum operations and concrete protocols wherever possible, rather than invoking quantum phenomena as functional solutions. Second, recognize that the doctrinal line between Hastings and Baughman is not yet settled, and that outcomes in quantum computing enablement cases may remain difficult to predict, at least until the Federal Circuit has an opportunity to harmonize the PTAB’s approach.

What About Nayfeh vs. Cao?

Nayfeh and Cao present a similar puzzle. In Nayfeh, the specification disclosed two specific examples (nitrogen vacancies in diamond and in silicon carbide), and the Board found the disclosure insufficient to enable the broader genus of “semiconductor vacancy qubit structures.” In Cao, the specification disclosed two specific objective functions (at paragraphs 49 and 57), and the Board found the disclosure adequate to support claims reciting a genus of objective functions satisfying certain mathematical constraints. Two examples were not enough in one case and were enough in the other. What explains the divergence?

As with Baughman vs. Hastings, the Nayfeh and Cao decisions can potentially be reconciled by pointing to differences in the nature of the claimed genus and the predictability of the underlying technology. The Nayfeh genus spanned a vast space of physical implementations: different semiconductor materials exhibit radically different crystal structures, defect formation energies, spin coherence times, and fabrication challenges, and a vacancy-based qubit in one material does not predict the behavior of a vacancy-based qubit in another. The two Nayfeh examples were specific solutions to specific materials-science problems, not illustrations of a transferable principle. The Cao genus, by contrast, was constrained by mathematical structure: any objective function meeting the claim limitations had to depend on specific, well-defined mathematical objects, and a skilled artisan in variational quantum algorithms could extrapolate from the two examples to understand how other functions meeting the same constraints would behave. On this reading, the number of examples matters less than whether the examples, together with the specification’s broader teaching, convey a general principle that skilled artisans can apply to implement other embodiments.

The practical implication may be that practitioners cannot rely on any numerical rule of thumb for example count. Two examples may suffice in mathematically or structurally constrained domains where the specification articulates a general principle connecting the examples. Two examples are likely insufficient in unpredictable physical-implementation domains where each embodiment requires separate experimental validation. When in doubt, drafters should either describe more species spanning the breadth of the genus, articulate the general principles that make the examples work, or narrow the claims to match the scope of what the specification genuinely enables. And as with Hastings and Baughman, practitioners should recognize that the doctrinal line here is not settled, and that outcomes may remain difficult to predict in close cases.

Practical Guidance for Practitioners

Based on these PTAB decisions, patent practitioners working on quantum computing applications should consider the following strategies:

1. Provide Extensive Description of Background Technologies

The PTAB decisions repeatedly emphasize that quantum computing is “esoteric technology” that may not be familiar to examiners or judges. Your specification should serve an educational function.

Best Practice: Include a comprehensive background section that explains:

• Fundamental quantum mechanics concepts relevant to your invention (superposition, entanglement, decoherence, interference)
• The quantum computing stack and where your invention fits
• How quantum and classical computing components interact
• The specific technical problem your invention solves and why quantum computing is necessary

The goal is to ensure that any examiner, judge, or jury member reading your specification could understand what the invention is and how it works, even without prior quantum computing expertise.

2. Describe Multiple Implementations with Connecting Principles

The Nayfeh decision demonstrates that describing two examples may not be sufficient for broad claims. However, Cao shows that you don’t need to describe every possible embodiment.

Best Practice:

• Describe several specific implementations spanning the breadth of your claims
• Explain the general principles that connect these implementations
• Provide guidance on how one skilled in the art could apply these principles to implement variations
• Consider whether your claims should be narrowed to match the scope of what you’ve truly enabled

For example, if claiming “a semiconductor vacancy qubit structure,” don’t just describe nitrogen vacancies in diamond and silicon carbide. Consider describing:

• The general principles of how vacancy-based qubits function
• What characteristics make a semiconductor suitable for vacancy qubits
• How vacancies are created and controlled in different materials
• The relationship between material properties and qubit performance

3. Avoid “Black Box” Quantum Features

The Baughman rejection shows the danger of invoking quantum computing concepts without implementation detail.

Best Practice: Never include quantum computing features as unexplained “magic.” For every quantum computing element in your claims:

• Explain the specific quantum hardware or technique involved
• Describe how it interfaces with classical computing components
• Detail the quantum operations performed and why they work
• Provide sufficient technical detail that a skilled artisan could implement it

For example, if you claim “using quantum entanglement,” explain:

• How the qubits are entangled
• What type of entanglement is created
• How the entangled state is maintained, at least in general terms
• How the entanglement is measured or used
• What quantum gates or operations are applied

4. Distinguish Classical and Quantum Operations Clearly

Many quantum computing inventions involve hybrid classical-quantum systems. The Cao case succeeded in part because it clearly delineated which operations occurred on classical vs. quantum hardware.

Best Practice:

• Explicitly state which steps are performed on classical computers and which on quantum computers
• Explain the interfaces and data flows between classical and quantum components
• Describe how classical data is encoded into quantum states and vice versa
• Detail any error correction or noise mitigation applied during the quantum-to-classical transition

5. Provide Mathematical and Technical Detail

Quantum computing is fundamentally mathematical. The Cao specification succeeded by providing specific mathematical formulations.

Best Practice:

• Include mathematical expressions for quantum operations, objective functions, and algorithms
• Explain the quantum mechanical basis for why your approach works
• Provide matrix representations of quantum gates where relevant
• Include equations showing how classical parameters relate to quantum observables

This detail serves multiple purposes: it demonstrates possession, provides enablement, and helps establish that your invention is a specific technical implementation rather than an abstract idea.

6. Address the Wands Factors Proactively

Examiners frequently apply the Wands factors when challenging enablement. Address these factors in your specification:

Quantity of experimentation: Minimize the experimentation needed by providing detailed protocols, parameters, and specifications.

Direction or guidance: Provide step-by-step implementation guidance, not just high-level descriptions.

Working examples: Include experimental results, simulations, or detailed worked examples demonstrating your invention.

Breadth of claims: Consider whether your broadest claims truly match the scope of enablement, or whether a hierarchy of claims from narrow to broad would be more defensible.

State of the prior art: Acknowledge the current state of quantum computing technology and explain what was known at your filing date.

Predictability: In unpredictable fields like quantum computing, more detail is required. Don’t assume that skilled artisans can easily extend your specific examples.

7. Claim Strategically Across the Technology Stack

Quantum computing innovations can occur at multiple levels of the technology stack. Consider where your invention sits and claim accordingly:

Layer 1 (Quantum Register): Claims to physical qubit implementations, materials, and fabrication methods. Typically requires detailed materials science and physics disclosure.

Layer 2 (Control Systems): Claims to enabling hardware, control signals, error mitigation at the hardware level. Focus on signal processing, timing, and interfacing with quantum hardware.

Layer 3 (Error Correction): Claims to algorithms mapping noisy physical qubits to stable logical qubits. Balance mathematical detail with practical implementation guidance.

Layer 4 (Software/Application): Claims to quantum algorithms and hybrid classical-quantum applications. Clearly distinguish quantum from classical operations, and demonstrate technical improvement rather than abstract mathematical concepts.

Best Practice: Consider whether “quantum”-specific limitations are necessary in independent claims or better reserved for dependent claims. Some inventions may have broader application beyond quantum computing, and claiming too narrowly may unnecessarily limit scope while also increasing the risk of enablement and insufficiency rejections.

Conclusion

Enablement remains a significant hurdle for quantum computing patent applications. The PTAB decisions in Nayfeh and Baughman demonstrate that even sophisticated applicants represented by experienced counsel can fail to satisfy § 112(a) requirements.

However, the success in Cao and Hastings shows that quantum computing inventions can be patented when properly disclosed. The key is providing sufficient technical detail, mathematical specificity, and implementation guidance to enable a person of ordinary skill in the art to make and use the full scope of the claimed invention – without undue experimentation and based on the knowledge available at the filing date.

As quantum computing technology matures and more patents enter prosecution and litigation, these early PTAB decisions will shape how future applications are examined. Practitioners who understand these enablement challenges and address them proactively in their specifications will be better positioned to secure valuable patent protection in this revolutionary field.

The quantum computing revolution is here. With careful attention to enablement requirements, patent practitioners can help ensure that breakthrough innovations receive the protection they deserve.

---

This article is based on a presentation by Matt Norwood, Patent Attorney at SLW, as part of the SLW Quantum Computing Webinar Miniseries. The complete webinar recording, “Patent Prosecution Strategies in Quantum Computing,”is available on the SLW Institute. The views expressed in the article do not constitute legal advice.