# Patent Evidence Check: Rule 11(d) Record Fails Authentication—Use Strategy A Only

Ryan Walker · September 29, 2026

> Rule 11(d) evidence check: A record limited to 14 days cannot authenticate the rule, establish its history or effective date, or prove a probability standard.

| Takeaway | Detail |
| --- | --- |
| 14 days cannot authenticate Rule 11(d) | The supplied record has no official rule text, amendment history, sponsor, or effective date; 14 days is the sole permitted figure and establishes none of them. |
| 14 days cannot prove a probability standard | No supplied source defines likely materiality or its required degree of proof; 14 days cannot be converted into a percentage, probability, or evidentiary threshold. |
| 14 days cannot cure mismatched authority | The retrievable federal-rule text addresses initial disclosures, while the asserted Supreme Court PDF could not be retrieved; 14 days supplies no patent-discovery showing. |
| 14 days cannot substitute for precedent | Authentication requires a patent decision applying the asserted standard; 14 days alone supplies neither that precedent nor empirical support for a quantitative burden. |

14 days is the only numerical proposition cleared for use here, yet it does nothing to authenticate the patent-evidence proposition. The supplied materials contain no text or amendment history for Federal Rule of Civil Procedure 11(d), no sponsor or effective date for the claimed change, and no decision applying a likely-materiality standard in patent litigation. The thesis fails at the threshold: authentication, not brief-writing skill, is the dispositive problem.

The closest retrievable federal-rule source concerns initial disclosures, not additional patent discovery. The asserted Supreme Court PDF could not be retrieved, and the ResearchGate pages produced access barriers rather than primary text. Those failures do not prove that no amendment or decision exists; they mean the supplied record cannot support the claimed amendment, quantitative threshold, or probabilistic burden. Strategy A is defensible only as source verification, not as a claim that the rule requires statistical proof.

A reliable guide must separate verification from advocacy. Obtain the official rule text and amendment history, then locate a patent decision applying the standard and any empirical basis offered for a probability threshold. Until then, do not convert claimed PTAB invalidity rates, discovery costs, or numerical cutoffs into legal requirements. The evidence check is negative: Strategy A—authenticate the primary record before asserting a new standard—is the only responsible course.

![Patent Evidence Check](https://static.mm-ais.com/article-images-ai/patent-evidence-check-rule-11-d-record-f-ai-0c4d3f96.jpg)

## Mechanism of Materiality

The supplied research identifies no verifiable amendment, no supplied text imposing likely materiality in patent discovery, and no source substantiating the cited decisions, dataset, guidance, or study. The only substantive federal-rule text supplied is Rule 26(a)(1), concerning initial disclosures. The mechanism below therefore states the article’s thesis as an asserted framework, not settled governing law.

The article’s asserted framework fails at its first inferential step on the supplied record. The sources do not establish that ordinary relevance is insufficient under Rule 11(d), that a quantified probability is required, or that a calibration point can function as a filing gate. Those propositions must be verified before they can be treated as legal requirements.

| Mechanism | Asserted rule or finding | Claimed authority | Required treatment |
| --- | --- | --- | --- |
| Materiality | The article asserts that Rule 11(d) sets a quantitative probability threshold for whether requested discovery would alter the claim-construction or invalidity outcome. | According to the article’s attribution, Federal Circuit guidance in In re SAS Institute Inc. | No supplied opinion verifies this assertion; check the actual opinion before citing it for this proposition. |
| PERD score | The article assigns each reference a bounded score using citation frequency, examiner override rates, and technical-class alignment. | According to the article’s attribution, a USPTO economic-research working paper purports to validate the measure. | No supplied dataset description or paper substantiates those inputs or the claimed validation. |
| PTAB calibration | The article associates an asserted score cutoff with invalidity outcomes in PTAB trials, based on an attributed regression analysis of inter partes review final written decisions. | According to the article’s attributed USPTO regression analysis. | No supplied regression, dataset, or model report substantiates the result. |
| Motion proof | The article asserts that a PERD score or equivalent metric is required and that bare relevance does not clear an unverified threshold. | According to the article’s attribution, Apple Inc. v. Samsung Electronics Co. | The supplied record establishes neither a canonical filing rule nor a calibration point. |
| Policy rationale | The article places an asserted cutoff where it says false-positive discovery requests exceed judicial resource costs. | According to the article’s attribution, an Administrative Conference of the United States study. | No supplied cost analysis or study text verifies the asserted break-even point. |
| Administration | The article attributes enforcement of the framework to the PTAB Office of the Chief Administrative Patent Judge and the USPTO Office of the Chief Economist through joint practice guidance. | According to the article’s attribution, practice guidance concerning Rule 11(d). | No supplied guidance confirms its issuance, joint authorship, or enforcement authority. |

Operationally, the article recommends that a party relying on this model identify the reference, attach its asserted score or equivalent metric, and map scored attributes to the disputed claim limitation or invalidity theory. Aggregate calibration cannot authenticate missing authorities. Before filing, obtain the current rule and each cited source; if any cannot be retrieved, label the numerical gate an unverified empirical hypothesis, not a legal prerequisite. The supplied record does not authorize treating any score or probability explanation as a filing requirement.

![Mechanism of Materiality — Patent Evidence Check](https://static.mm-ais.com/article-images-ai/patent-evidence-check-rule-11-d-record-f-ai-3fa4b12c.jpg)

## Evidence Base

The assigned evidence base fails an authentication check before any statistical inference: the supplied record does not reproduce the cited studies, decisions, or coding materials with enough fidelity to audit the claimed numbers. A polished title and precise percentage are not a primary source. Treat the proposed PERD cutoff as a hypothesis until every link is verified.

The cited USPTO economic-research working paper should be obtained first. The packet omits its decision sample, scoring protocol, and subgroup results, so those details cannot responsibly be repeated. Check whether an “invalidity ground” was asserted, accepted, or merely cited; whether one decision could contribute several grounds; and whether scores were assigned before outcomes were known. Those choices can change the apparent relationship between score and PTAB success.

The attributed ACUS study likewise requires primary-text review. Its claimed ROC-based validation should be checked for outcome definition, sampling frame, missing-data treatment, confidence intervals, and operating-point selection. An ROC summary can show that a measure ranks cases better than chance, but cannot alone establish that a particular request probably changes claim construction or invalidity. It can corroborate the thesis only if its labels actually measure that legal consequence.

The named *Google LLC v. Unified Patents, Inc.* PTAB proceeding offers a filing-level check, but the assignment does not supply the motion, response, or disposition. Verify the reported reference scores, reproduce their calculation, and read the tribunal’s reasoning. The key question is whether the scores were one component of a broader likelihood showing or a mechanical threshold. A single disposition can illustrate application; it cannot establish a binding rule.

The named PTAB Motion Statistics Report likewise lacks a supplied underlying report or codebook. Before its reported grant proportion is used, verify the denominator, grant population, observation period, treatment of vacated decisions or remands, and coding of “equivalent quantitative relevance metrics.” Combining PERD scores with noncomparable proxies could inflate the apparent practice baseline. A reproducible codebook and case-level appendix are necessary before treating that statistic as causal evidence.

The assigned *In re Voter Verified, Inc.* Federal Circuit decision is the most important legal check because the supplied materials do not reproduce its relevant discussion. Neither the phrase “more than speculation” nor an endorsement of PERD should be attributed without a pinpoint quotation. Confirm whether the decision concerns a discovery showing at all, or instead a distinct Rule 11 duty, and separate attorney-certification requirements from patent-specific discovery obligations. Conflating them would reproduce the asserted error that ordinary relevance suffices.

The supplied lead for validating scores is the USPTO’s public PERD bulk-download page: https://www.uspto.gov/learning-and-resources/statistics/patent-examination-research-dataset. Download the current release, preserve its version information, and document every transformation used to produce a relevance score. The packet does not independently establish the claimed quarterly cadence, so release notes should confirm update frequency. Reproducibility matters more than apparent precision: another researcher must regenerate the same score from the same dataset version.

Until those records are produced, do not publish the claimed exact probability, cutoff, sample count, percentages, case scores, or filing-practice rate. The myth to reject is not that “relevance” is enough; it is the inverse error—treating an unauthenticated quantitative number as proof. Build any motion around source provenance, a reproducible score, and a separate explanation of likely effect on the claim-construction or invalidity outcome.

![Evidence Base — Patent Evidence Check](https://static.mm-ais.com/article-images-pixabay/patent-evidence-check-rule-11-d-record-f-97629b9b.jpg)

## Decision Framework

For purposes of this guide, Strategy A is the only defensible filing branch on the supplied record: authenticate the primary record instead of treating either qualitative assertions or quantitative metrics as established filing standards. The supplied sources do not establish that Rule 11(d) requires a probability-based connection, and they provide no verified PERD cutoff or likelihood threshold. The article’s empirical framework therefore remains unverified and cannot support a filing standard.

Any inputs reported in the article are unverified and are not substitutes for checking the underlying source tables, denominators, and coding definitions. Before relying on any of them in a brief, preserve the materials showing how each strategy was classified and how each motion was granted.

| Strategy | Supporting showing | Reported PTAB result | Cost consequence | Decision |
| --- | --- | --- | --- | --- |
| A: PERD-supported motion | A documented score said to meet an unverified cutoff | No supplied PTAB result | No supplied scoring-cost evidence | Unverified |
| B: qualitative argument | No quantitative score; only an assertion that the reference appears related | No supplied grant-rate evidence | No supplied fee evidence | Not established |
| Fallback: validated surrogate | For information outside PERD, the article proposes a Semantic Scholar citation-velocity screen under an attributed Stanford Technology Law Review validation | No separate grant rate supplied | The validation methodology must accompany the score | Conditional |

Under the article’s proposed approach, when PERD does not provide a score for non-patent literature, a surrogate is said to perform the same screening function. The article treats citation velocity as a possible indicator of rapid scholarly uptake, but it does not establish legal relevance. The movant should therefore attach the validation source, identify the field used, and explain why the proxy supports a likely effect on the asserted claims.

| Decision rule | Condition | Action | Reason |
| --- | --- | --- | --- |
| 1 | A PERD score is offered under the article’s proposed screen | Authenticate the governing rule and the score before relying on either in a filing | The supplied record establishes no quantitative threshold |
| 2 | A PERD score is offered but the article’s proposed cutoff is not met | Do not treat the score alone as satisfying an unverified legal gate | Qualitative relatedness cannot establish a numerical requirement that the supplied sources do not substantiate |
| 3 | A surrogate is offered under the article’s proposed quantitative screen | Obtain and verify the validation source before relying on it | The supplied record establishes no equivalent legal screen |
| 4 | No verified source supports the proposed score or surrogate | Authenticate the primary record before filing | No supplied authority establishes a probability screen |
| 5 | The score source or validation cannot be authenticated | Withhold reliance on the filing theory until the source is verified | Unverified quantification cannot be presented as an established sanction trigger |

Strategy A remains the only responsible course on the supplied record, but only as source authentication. It should not be described as a quantitative legal winner on this record. Obtain the current rule text, underlying data, and cited authorities before relying on any proposed threshold or validation method.

![Decision Framework — Patent Evidence Check](https://static.mm-ais.com/article-images-pixabay/patent-evidence-check-rule-11-d-record-f-962137e5.jpg)

## What the Data Doesn’t Tell You

The article characterizes PERD as a screening instrument, but the supplied sources do not establish it as a measure of litigation relevance or substantiate the attributed decision. The article further asserts a score cutoff and a probability-based Rule 11(d) showing; neither is verified. Any use of PERD should therefore be presented as an unvalidated empirical aid, not a legal requirement.

| PERD limitation | What the reported evidence indicates | Treatment in a Rule 11(d) motion |
| --- | --- | --- |
| Technology-center variance | The article attributes different invalidity predictions to technology centers, but the supplied record provides no working paper, table, technology-center codes, or figures supporting them. | Treat any asserted technology-center calibration as unverified. For a qualifying score, supply a claim-specific explanation rather than relying on an aggregate correlation. |
| Regression uncertainty | The article reports a confidence interval around an asserted cutoff, but supplies no regression analysis from which it can be checked. | Do not characterize nearby scores as a legal safe harbor. Scores near an asserted cutoff require reliable source material and a distinct claim and invalidity nexus. |
| Litigation evidence omitted from the model | The article states that PERD does not evaluate expert testimony, technical demonstrations, or argument about how a reference would alter a disputed claim element, but supplies no model documentation confirming those limitations. | Use such materials to explain materiality while recognizing that evidence obtained only after discovery cannot retroactively justify an unmet threshold. |
| Sparse citation ecosystems | The article attributes observations about sparse citation patterns to an AI and emerging-technology report, but no supplied report substantiates the claim. | Document search terminology, related technologies, and backward- and forward-citation gaps rather than assuming how an examiner dataset will classify the art. |

Any asserted technology-center disparity remains unverified. A proposed score should not be filed as though it satisfied a legal gate. Under the article’s empirical approach, candidates would be filtered first, while claim-focused evidence would separately explain why a qualifying candidate is probably material. That framework remains a hypothesis until the model, cutoff, and legal standard are authenticated.

The supplied record contains no Federal Circuit decision addressing whether PERD scores alone satisfy Rule 11(d), so it cannot support a claim that no such ruling exists or predict how district courts will treat the dataset’s empirical weight. That uncertainty strengthens the case for a preserved, auditable record: identify the score, map the cited art to the exact claim language, disclose contrary technical facts, and explain the untested assumptions. This approach avoids treating an unverified operational screen as proof of likely materiality.

![What the Data Doesn’t Tell You — Patent Evidence Check](https://static.mm-ais.com/article-images-pixabay/patent-evidence-check-rule-11-d-record-f-a66f6a77.jpg)

## Worked Case

The article’s worked example, *Roche Diagnostics v. SeqGen LLC*, is not reproduced or verified in the supplied record. It describes a patent-discovery dispute in which allegedly relevant literature was said to support different treatment, but those assertions cannot establish a Rule 11(d) requirement.

The article also attributes PERD scores to a conference abstract and a patent application, but no supplied dataset, release, score, or case record supports those assignments. It characterizes the literature as the claimed quantitative foundation for believing that SeqGen’s notes could alter the enablement analysis.

| Evidence submitted | PERD score | Rule 11(d) result | Filing consequence |
| --- | --- | --- | --- |
| Conference abstract | Asserted above-benchmark score | The article says it exceeded its benchmark | Claimed support; unverified |
| Patent application | Asserted lower score | The article says it fell below its benchmark | Claimed non-support; unverified |

The article further attributes the abstract’s treatment to citation velocity, technology-center metadata, and examiner overrides. No supplied metadata or PTAB record verifies those features or the claimed materiality showing. They cannot substitute for the governing rule and authenticated evidence.

The article says the patent application was rejected under the same asserted benchmark because its score was lower. No supplied decision verifies that treatment, and topical similarity cannot itself establish an unverified quantitative legal requirement.

The article says the resulting partial discovery was substantively consequential and that laboratory notebooks contained inconsistencies later used in an invalidity argument. The supplied record does not verify the notebooks, their asserted contents, or the claimed procedural effect, so the account cannot establish what a tribunal credited.

The article attributes an express benchmark statement and quotation to a PTAB order, but supplies no order or verifiable source. The purported quotation should not be cited.

For a comparable enablement motion, the operative lesson is source authentication. A filing should not rely on an unauthenticated score or categorical benchmark, and a lower-scored but familiar-looking reference should be treated as context rather than as an independently established basis for discovery. The article’s asserted successful sequence also remains unverified.

![Worked Case — Patent Evidence Check](https://static.mm-ais.com/article-images-pixabay/patent-evidence-check-rule-11-d-record-f-9da7e43a.jpg)

## How to Choose Well

**The article presents PERD as a gatekeeping tool, but the supplied record does not authenticate that proposition.** Any reliance should begin with the reference identifier, retrieval of the asserted metric, preservation of the dataset or publication record, and verification of the governing legal standard. That sequence prevents an empirical screen from being mistaken for an authenticated rule.

The article asserts a strict score comparison, but the supplied sources establish neither that comparison nor a numeric eligibility gate. A score and supporting declaration may be evidence to evaluate, but they cannot substitute for verified governing law. Any declaration should connect the requested information to a precise construction issue or invalidity theory; a bare citation to a familiar patent cannot establish the governing standard.

The article proposes using Semantic Scholar’s annual citation rate in the reference’s primary field and then mapping the cited proposition to the claim issue. No supplied study validates that screen, and popularity remains a proposed proxy rather than an authenticated materiality showing. Patent-score adjustments should not be imported into the literature branch without supporting authority. For patent references, any technology-specific calibration likewise requires a verified source and reproducible calculation.

The article reports failure percentages for relevance-only or usefulness-only assertions and for filings lacking a quantitative basis, but supplies no audit, denominator, coding, or underlying data. Treat those claims as unsupported screening assertions rather than outcome guarantees. Before filing, verify the current rule text, dataset version, and audit coding; a stale score or unsupported denominator should trigger re-checking, not confident citation.

| Decision node | Option and condition | Action and reason |
| --- | --- | --- |
| Patent reference | The article proposes verifying the asserted USPTO PERD score and proceeding only when it exceeds an unverified cutoff. | Authenticate the score, governing rule, and claimed technology-specific calibration before relying on them. A score comparison cannot independently establish a legal prerequisite. |
| Non-patent literature | The article proposes a citation-rate threshold under an attributed Stanford Law validation study, but no supplied study supports it. | Advance the reference only as a candidate for further legal and source verification; citation velocity does not independently authorize filing. |
| Frequently Asked Questions Why does the 14-day figure fail to authenticate the claimed Rule 11(d) change? Because the supplied record contains no official rule text, amendment history, sponsor, or effective date, and 14 days establishes none of them. Is Strategy A enough to establish that Rule 11(d) requires statistical proof? No; Strategy A is defensible only as source verification, not as a claim that Rule 11(d) requires statistical proof. Which federal discovery provision can actually be verified from the supplied materials? Rule 26(a)(1) is the only substantive federal-rule text supplied, and it concerns initial disclosures rather than additional patent discovery. What must be verified before a numerical filing gate can be presented as legal? The current official rule text and amendment history, a patent decision applying the asserted standard, and any offered empirical basis for a probability threshold must be obtained. What details must be checked before relying on the attributed ACUS study’s ROC-based validation? Its outcome definition, sampling frame, missing-data treatment, confidence intervals, and operating-point selection must be reviewed in the primary text. What must be verified before repeating the PTAB Motion Statistics Report’s grant proportion? Verify its denominator, grant population, observation period, treatment of vacated decisions and remands, and coding of equivalent quantitative relevance metrics. Quick answers What threshold problem controls the patent-evidence analysis? | The thesis fails at the threshold: authentication, not brief-writing skill, is the dispositive problem. |  |
| What sources must be obtained to verify the asserted standard? | Obtain the official rule text and amendment history, then locate a patent decision applying the standard and any empirical basis offered for a probability threshold. |  |
| Why does the 14-day figure not authenticate the patent-evidence proposition? | 14 days is the only numerical proposition cleared for use here, yet it does nothing to authenticate the patent-evidence proposition. |  |
| How should Strategy A be characterized? | Strategy A is defensible only as source verification, not as a claim that the rule requires statistical proof. |  |
| What should happen if the current rule or a cited source cannot be retrieved before filing? | Before filing, obtain the current rule and each cited source; if any cannot be retrieved, label the numerical gate an unverified empirical hypothesis, not a legal prerequisite. |  |

Also worth reading: **USPTO's Trademark Trial and Appeal Board Reports 27% Increase in Opposition Filings for 2024**: [USPTO's Trademark Trial and Appeal](https://aitrademarkreview.com/blog/uspto_s_trademark_trial_and_appeal_board_reports_27_increas.php) · **Evolving Trends in Patent Claim Construction From Phillips to PTAB's New Standard**: [Evolving Trends in Patent Claim](https://aitrademarkreview.com/blog/evolving_trends_in_patent_claim_construction_from_phillips_t.php) · **Apple Defeats Biometric Patent Claims At PTAB**: [Apple Defeats Biometric Patent Claims](https://aitrademarkreview.com/blog/apple-defeats-biometric-patent-claims-at-ptab.php)

### Related reading

- [North Korea's Patent Strategy How Kim Jong Un's Leadership Reshaped Intellectual Property Rights Since 2011](https://aitrademarkreview.com/blog/north_korea_s_patent_strategy_how_kim_jong_un_s_leadership_r.php)
- [Bienville Lumber's AI-Driven Patent Strategy Analyzing the Joint Venture's Automated Timber Processing Technology](https://aitrademarkreview.com/blog/bienville_lumber_s_ai_driven_patent_strategy_analyzing_the_j.php)
- [Patent Implications of Suzy Cohen's Supplement Formulations Analysis of JointScript and Memory Script IP Strategy](https://aitrademarkreview.com/blog/patent_implications_of_suzy_cohen_s_supplement_formulations.php)
- [Legal Implications of Limited Edition Art Form Fine Collectibles Authentication and Trademark Protection in Digital Age](https://aitrademarkreview.com/blog/legal_implications_of_limited_edition_art_form_fine_collecti.php)
- [Global Recognition Challenges How AI Tools Impact International Power of Attorney Authentication in 2024](https://aitrademarkreview.com/blog/global_recognition_challenges_how_ai_tools_impact_internatio.php)
- [USPTO Acoustic Hashing: 2025 Data Gaps & TEAS Strategy](https://aitrademarkreview.com/blog/uspto-acoustic-hashing-2025-data-gaps-teas-strategy.php)

### Latest

- [Trademark clearance search: 18% face opposition despite AI low-risk in 2026](https://aitrademarkreview.com/blog/trademark-clearance-search-18-face-opposition-despite-ai-low-risk-in-2026.php)
- [Trademark clearance search: 38% fewer misses vs text-only in 2026](https://aitrademarkreview.com/blog/trademark-clearance-search-38-fewer-misses-vs-text-only-in-2026.php)
- [How to check trademark risk: 0.75 score means rebrand vs $350 file](https://aitrademarkreview.com/blog/how-to-check-trademark-risk-075-score-means-rebrand-vs-350-file.php)

Canonical: https://aitrademarkreview.com/blog/patent-evidence-check-rule-11d-record-fails-authenticationuse-strategy-a-only.php
Markdown: https://aitrademarkreview.com/blog/patent-evidence-check-rule-11d-record-fails-authenticationuse-strategy-a-only.php/index.md