Check A Pursuit™ · Analysis Report
Example Check A Pursuit™ Analysis Report
A public example of a Check A Pursuit™ report, prepared by Patent Attorney John Ogilvie, based on a fictional unfiled draft provisional patent application on an interactive holographic environment simulation system.
This is a public example of a Check A Pursuit™ Analysis Report. It is prepared from a version of a fictional unfiled draft provisional patent application on an interactive holographic environment simulation system. The vulnerabilities, discovery requests, and leverage strategies identified are real categories used in actual analyses. The analysis illustrates how AI-use indicia found in the document can support adversarial strategies in discovery, negotiation, and pre-filing advisory contexts.
This public example is not confidential client information. It illustrates the type of privileged attorney-directed analysis that may be provided in an actual client matter.
Your actual Check A Pursuit™ report will be tailored to your specific document, proceeding type, and situation.
Check A Pursuit™ Analysis Report
Prepared by Patent Attorney John Ogilvie · Ogilvie Law Firm
Interactive Holographic Environment Simulation System
Fictional Draft Provisional Patent Application — public example
This Check A Pursuit™ Analysis Report was prepared by Registered Patent Attorney John Ogilvie of Ogilvie Law Firm as part of a limited scope legal representation to provide legal advice. Although portions of this report were generated by artificial intelligence (AI), the AI output was scrutinized and edited by Attorney Ogilvie and reflects his opinions, mental impressions, and legal judgment.
This report identifies conduct-based vulnerabilities in the patent and prosecution history arising from the behavior of the inventor, patent drafter, or patent prosecutor — specifically, indicia of AI use during conception, drafting, and prosecution, and the adversarial or negotiation strategies those indicia support. This report addresses a conduct-based dimension that is outside the scope of a Check A Patent™ structural analysis.
All findings are framed in terms of how an adverse party would use them. This framing serves both an adverse client seeking to challenge or negotiate the patent and a patent owner conducting pre-litigation exposure assessment. How the client uses this report is outside the scope of this engagement.
Proceeding type: This document targets an unfiled draft provisional; no proceeding is currently pending. The Leverage section is framed in Negotiation mode — exposure assessment describing what a litigation adversary could do with each finding if this application were filed and the patent were asserted. This framing is appropriate for pre-filing advisory use. If this application is later filed and becomes the subject of litigation, a new Check A Pursuit™ report should be generated at that time using the applicable proceeding type.
Findings (Detect)
Findings are ordered from most to least significant AI-use indicum.
Embedded Workflow Artifacts — AI Template Placeholders and Draft Label
Significance: Very High
What Was Found:
The cover block of the document contains multiple AI-generation artifacts that would survive into any filing made from this draft if not corrected. These patterns are characteristic of a document-generating AI that was prompted to produce a complete patent application and generated a structured cover block with field labels but populated only those fields it could complete from the prompt — leaving all inventor-specific and filing-specific fields as bracketed placeholders.
No professionally drafted provisional patent application submitted by an attorney intentionally carries bracketed unfilled placeholder fields in the cover block. The presence of these artifacts, combined with the explicit draft label, is strongly consistent with AI generation of a complete application document in a single pass from a structured prompt that did not supply the full set of required field values.
If this document were filed with these artifacts intact, the placeholder text would appear in the USPTO’s public record.
Analysis Text:
The document’s cover block contains bracketed placeholder fields of the form [INVENTOR NAME], [FILING DATE], [APPLICATION NUMBER], and similar. These are unfilled template slots, not redactions. Co-occurrence with the explicit “DRAFT FOR ATTORNEY REVIEW” label in the document header substantially strengthens the workflow-artifact inference: the document was generated as a complete draft artifact and was not subsequently cleaned for filing or distribution.
→ Supports Seek requests: S-1, S-3 | Leverage analysis: L-4 (Authorship-Bridging)
AI Prior Art Gap Analysis Embedded in Background
Significance: High
What Was Found:
The Background section of this application contains a formal prior art gap analysis structured as a statement of unmet need. This passage is not a description of a technical problem in the inventor’s field. It is a formal enumeration of what the prior art cannot do — phrased as “need for” and “limited in their ability to” — that tracks the claim structure of the application directly. The three needs identified (physically interactive objects, physical resistance and handling, unbounded simulation from a fixed room) map precisely to independent claim 1’s three subsystems (projection, force field, matter replication) and claim 16’s spatial expansion mechanism.
A human inventor describing the background of a simulation system would describe specific limitations of known systems they encountered or researched. The document-generating AI, prompted to write a patent application, apparently generated a Background section that argues the prior art does not disclose the specific claim elements — which is a legal assessment of patentability, not a description of the technical problem.
If this application were filed and the patent were asserted, a litigation adversary could use this Background passage to: (a) establish that the applicant (through their AI tool) was aware of or assessed the prior art landscape; (b) probe whether specific prior art references evaluated by the AI were disclosed on an Information Disclosure Statement; and (c) if the AI-assessed prior art landscape included material references that were not disclosed, support a 37 C.F.R. § 1.56 duty-of-candor argument as to those specific references.
Analysis Text:
Background paragraph states: “There exists a need for simulation systems capable of producing fully immersive, physically interactive environments in which users can not only observe a simulated setting but physically engage with objects, characters, and environments within it. Conventional simulation approaches are limited in their ability to produce objects that feel physically solid and can be handled, manipulated, or resisted by a user. There is also a need for simulation systems that can present an apparently unbounded environment to a user who is physically confined within a fixed-size room.”
This language frames the invention as solving a problem specific to users confined in a fixed-size room with physically solid objects. The three enumerated “needs” map claim element by claim element to independent claim 1 (physically interactive objects, physical resistance, handling) and claim 16 (unbounded simulation from a fixed room). The structural alignment between the Background’s gap analysis and the claim architecture is an indicator of AI-generated patent applications.
→ Supports Seek requests: S-1, S-3, S-IDS | Leverage analysis: L-2 (AI Duty of Candor), L-3 (AI Prior Art)
AI Stylistic Markers — Four Co-Occurring Sub-Triggers
Significance: High
What Was Found:
Four independently sufficient AI stylistic markers are present in the specification, co-occurring in a single document. Co-occurrence of multiple sub-triggers substantially strengthens the overall AI-usage likelihood.
This is a public example document. Your actual Check A Pursuit™ report will list AI stylistic markers detected in your particular document. Although some of these markers sometimes occur in documents unrelated to patents, the Check A Pursuit™ analysis is tailored to identify and leverage AI usage markers that often occur in AI-generated patent applications or USPTO Office Action responses.
A litigation adversary could use these findings to support the AI authorship narrative across all deposition and cross-examination strategies, and as a basis for requesting production of AI session transcripts.
Analysis Text:
Each sub-trigger is independently sufficient to support an AI-use inference. Co-occurrence across four distinct categories — formatting convention, structural generation pattern, self-referential linguistic tell, and boilerplate reproduction — substantially raises the aggregate confidence level. No single trigger establishes AI use, but the four-trigger co-occurrence makes alternative explanations much less plausible.
→ Supports Seek requests: S-1, S-2, S-3 | Leverage analysis: L-4 (Authorship-Bridging), L-5 (Attorney Direction)
Excess Claims for a Provisional Application
Significance: High
What Was Found:
This provisional application contains 20 numbered claims. Provisional patent applications are not required to include claims under 35 U.S.C. § 111(b), and professionally drafted provisionals typically either omit claims entirely or include a small number to establish a disclosure baseline for a future non-provisional. A 20-claim provisional — structured with four independent claims of different statutory types and sixteen dependent claims — is characteristic of a document-generating AI prompted to “write a patent application” that generated a full claim set as part of its output, without any instruction limiting the claim set to provisional-appropriate scope.
The 20-claim structure is not itself a provisional filing defect. Its significance is that a full non-provisional-style claim set in a provisional draft may support an inference of AI- or template-assisted application generation and may create a record that later invites inquiry if filed.
The claim set is non-provisional quality in its scope and structure: it includes apparatus claims, a method claim, a safety-system-focused apparatus claim, and a spatial-expansion-focused apparatus claim — a multi-claim-type architecture more consistent with a non-provisional claim strategy than a provisional disclosure baseline.
Analysis Text:
The dependency map is: claim 1 (independent; 7 dependents: 2–8); claim 9 (independent method; 3 dependents: 10–12); claim 13 (independent system; 2 dependents: 14–15); claim 16 (independent system; 4 dependents: 17–20). Total: 4 independent claims + 16 dependent claims = 20 claims. For a provisional application, this is an unusually complete claim set. The multi-type claim architecture (apparatus ×3, method ×1) covering distinct aspects of the invention suggests a claim strategy document, not a disclosure baseline.
→ Supports Seek requests: S-1, S-3 | Leverage analysis: L-1 (AI Inventorship), L-4 (Authorship-Bridging)
Missing Claim Types — No CRM Claims for a Software-Orchestrated Architecture
Significance: High
What Was Found:
The claim set includes apparatus claims and a method claim but contains no computer-readable medium (CRM) claims and no software-implementation claims. The specification describes an invention centered on a central processing system that coordinates three hardware subsystems in real time, maintains simulation state, processes voice commands, responds to user physical interactions, and manages on-the-fly simulation construction from verbal descriptions. These are software-implemented functions that would normally support CRM claims.
The complete absence of CRM claims for a system in which software orchestration is the central inventive coordination mechanism is atypical of a strategically drafted claim set and consistent with AI generation without a claim-type coverage checklist.
Analysis Text:
The central processing system is described as performing: real-time coordination of three subsystems; simulation state management (position, appearance, physical properties, and state of every simulated object); voice command processing; user interaction response; and on-the-fly simulation construction from verbal descriptions. Each of these functions is software-implemented and would support a CRM claim covering the software that implements it. A strategically drafted claim set for this invention would include at least one CRM claim. The absence is a claim-type coverage gap consistent with AI single-pass generation.
→ Supports Seek requests: S-1, S-3 | Leverage analysis: L-1 (AI Inventorship), L-4 (Authorship-Bridging)
Unbalanced Dependent Claim Distribution
Significance: Medium
What Was Found:
The claim dependency tree shows a ratio of dependents under claim 1 to dependents under claim 13 of 7:2 — 3.5:1, marginally exceeding the 3:1 threshold that triggers this row. While claim 1 is the broadest independent claim and some concentration of dependents is expected, the marginal over-concentration is consistent with a single-pass AI generation pattern in which the AI exhausts dependent claim ideas under the first independent claim before addressing subsequent ones.
Analysis Text:
Claim 1: 7 dependents (claims 2–8). Claim 9: 3 dependents (claims 10–12). Claim 13: 2 dependents (claims 14–15). Claim 16: 4 dependents (claims 17–20). Ratio of claim 1 dependents to claim 13 dependents: 7:2 = 3.5:1. This finding is at Medium significance given the marginal threshold exceedance. Co-occurrence with F-4 and F-5 provides supporting context.
→ Supports Seek requests: S-1 | Leverage analysis: L-4 (supporting)
Legal Terminology in Specification Body
Significance: Medium-High
What Was Found:
The specification contains legal patent vocabulary in the specification body in contexts indicating the document-generating AI was prompted to produce a patent application. The Detailed Description opens with the verbatim boilerplate: “The following description of preferred embodiments is illustrative and not limiting; the scope of the invention is defined by the claims.” The phrase “preferred embodiment” and “preferred embodiments” appears multiple times in the Detailed Description. These usages indicate the document-generating AI was prompted to write a patent application and reproduced standard patent specification language from its training data.
Analysis Text:
The verbatim boilerplate disclaimer as the exact first sentence of the Detailed Description is an arguable indication of AI reproduction. “Preferred embodiment” language in the specification body is a standard patent drafting convention that AIs reproduce from training data. The verbatim, unmodified reproduction of this boilerplate is a tell: a human drafter would typically vary or personalize the language.
→ Supports Seek requests: S-1 | Leverage analysis: L-4 (supporting)
Specification Missing Drawing References for a Multi-Subsystem Hardware Invention
Significance: Medium
What Was Found:
The specification contains no drawings, no “Brief Description of the Drawings” section, and no reference numeral callouts anywhere in the Detailed Description. The invention — a multi-room hardware system with three distinct physical subsystems, a central processing system, spatial expansion hardware, a dual-modality control interface, and a safety protocol system — is a complex multi-subsystem hardware architecture. Provisional applications for complex hardware systems typically include at least a system block diagram or schematic. The complete absence of any drawing content or even a placeholder reference to drawings to be supplied before filing is consistent with AI generation of the specification as a text-only document from a prompt that did not include any figures or design materials.
Analysis Text:
Claim 1 requires five coordinated structural elements. Claim 9 requires a method of coordinating three subsystems in response to user interactions. Claim 13 requires a dual-authorization safety protocol system. Claim 16 requires a spatial expansion mechanism that adjusts the simulated environment in response to physical user movement. None of these architectures is depicted in any figure. The absence of drawings makes it harder for the patent owner in future proceedings to distinguish cited references based on structural distinctions not adequately described in the specification text.
→ Supports Seek requests: S-1 | Leverage analysis: L-4 (supporting)
Seek
Seek entries are structured as due diligence requests or as discovery requests that a litigation adversary would serve. Check A Pursuit™ does not include drafting discovery; that would be a separate engagement.
General AI Tool Identification
Identify all AI tools, platforms, services, and software used in the preparation of this provisional patent application, including from first draft through filing. For each tool identified, state the dates of use, the tasks performed, and whether use was directed by a licensed patent attorney or agent.
Serves: F-1, F-2, F-3, F-4, F-5, F-6, F-7, F-8
Platform Confidentiality and Training Settings
For each AI tool or platform identified in response to S-1: (a) identify the account type (personal, enterprise, API); (b) state whether a model training opt-out was in effect during the period of use and produce any documentation; (c) identify who had access to the account and any shared sessions; (d) produce the platform’s applicable terms of service as of the date of use.
Serves: F-1, F-3
AI Session Log Production
Produce all AI session transcripts, chat logs, prompt histories, and AI outputs generated in connection with the preparation of this provisional patent application. If production is objected to on privilege grounds, produce a privilege log identifying each withheld session by date, platform, user, and claimed privilege basis, sufficiently detailed to enable assessment of whether the privilege claim is sustainable.
Serves: F-1, F-2, F-3, F-4, F-5
Subject Matter Waiver Notice
Subject matter waiver notice: To the extent patent owner relies on produced AI outputs to support any position in this proceeding while withholding other AI outputs on the same subject matter, adverse party will seek a ruling of subject matter waiver under applicable evidentiary rules on the basis that the disclosed and withheld materials ought in fairness to be considered together — in particular, where selective disclosure would result in a misleading partial presentation.
Note: Subject matter waiver under applicable evidentiary rules requires intentional disclosure, same subject matter, and satisfaction of the fairness prong; it is not automatic on voluntary production.
Serves: All findings
AI Prior Art Production
Produce all AI outputs identifying, assessing, or characterizing prior art or prior art gaps generated during preparation of this provisional patent application, including but not limited to outputs resulting from prompts asking the AI to evaluate novelty, assess obviousness, identify competing references, or identify gaps in the prior art landscape for simulation systems with physically interactive environments. For each prior art reference or prior art gap identified by the AI, state whether that reference has been disclosed in an Information Disclosure Statement and provide the IDS entry by date. For each AI-identified reference not disclosed, explain the basis for the materiality determination and identify who made it and when.
Note: This request targets specific prior art references surfaced by the AI and available to humans — not the AI’s internal search methodology. The duty of candor under 37 C.F.R. § 1.56, if implicated, applies to material information known to human actors, not to AI session logs standing alone. If a human actor reviewed or possessed AI outputs identifying specific material prior-art references, those references should be assessed under the duty of disclosure at the appropriate non-provisional stage.
Serves: F-2
Leverage
Proceeding type: Pre-Filing Advisory (Negotiation framing). This application has not been filed. The Leverage analysis below is framed as exposure assessment — each finding is described in terms of what a litigation adversary could do with it if this application were filed and a patent based on it were later asserted. A patent owner reviewing this draft before filing should treat each L-entry as a fixable or unfixable vulnerability that will exist in the filed record if not corrected before filing.
Legal Caution: The AI-use indicators identified in the Findings section support targeted discovery and deposition questions concerning human conception, attorney review, claim drafting, and compliance with USPTO filing obligations. They do not, standing alone, establish invalidity, unenforceability, inequitable conduct, privilege waiver, or discoverability of AI sessions. Any legal consequence depends on the evidence developed through discovery and the governing law of the forum.
Current Guidance Note: USPTO guidance on AI-assisted inventions was revised on November 26, 2025, rescinding the February 2024 guidance in its entirety. Under current guidance, no separate inventorship standard applies to AI-assisted inventions. The same legal standard governs regardless of AI use: only natural persons may be named inventors, and AI systems are treated as tools that may assist human inventors. The Validity Attacks below use AI-use indicators to identify targeted discovery and deposition questions concerning human conception — they do not assert that AI assistance alone establishes an inventorship defect or patent invalidity.
Validity Attacks
AI Inventorship Forced-Choice Trap
A litigation adversary could serve the following discovery request: “For each independent claim of [patent number] — specifically claims 1, 9, 13, and 16 — identify the human inventor who conceived each claim and describe the conception event, including the date, the specific mental formulation of each claimed element, and the inventor’s contribution distinct from any AI tool used.”
Horn 1 — AI Inventorship Concession
If discovery reveals that the four-independent-claim structure of this application emerged from AI generation without specific human conception events that can be identified and described for each claim, the record may support an incorrect-inventorship challenge under applicable inventorship doctrine and current USPTO guidance.
Horn 2 — Obviousness Evidentiary Predicate
If the patent owner contends the AI merely drafted what the inventor independently conceived and described, testimony to that effect — combined with prior art evidence showing that the individual subsystems and their coordination are known, predictable, or technically conventional — creates an evidentiary predicate for an obviousness narrative under 35 U.S.C. § 103.
Pre-Filing Advisory Implication
If the inventor cannot describe a specific conception event for each independent claim element independent of the AI tool’s output, that gap should be addressed before filing — either through inventor documentation of conception or through attorney-guided claim drafting that is demonstrably tied to human-conceived elements.
Disclosure Obligation Attacks
AI Duty of Candor Forced-Choice Trap
A litigation adversary could serve the following discovery request: “Did any individual associated with the preparation or filing of [application number] use any AI tool to assess prior art, evaluate novelty or non-obviousness, or identify prior art gaps for simulation systems with physically interactive environments?”
Horn 1 — Yes
The Background section’s formal prior art gap analysis is evidence that the AI assessed the prior art landscape. If the AI surfaced specific material prior art references that were not disclosed on an IDS, that is the factual predicate for a 37 C.F.R. § 1.56 duty-of-candor analysis as to those specific references. The adversary could follow immediately with: “Produce all AI outputs identifying or assessing prior art generated during preparation of this application.”
Horn 2 — No
The Background section’s formal prior art gap analysis — structured as an enumeration of what existing simulation systems cannot do — may be difficult to explain as an inventor-authored technical problem description. Either the AI assessment occurred and is being concealed, or the AI was asked to generate a patent application without prior art context — which leaves the Background’s gap analysis language in the document without any corresponding materiality analysis by a human actor.
Inequitable conduct calibration: This trap supports discovery into materiality and intent. Inequitable conduct requires both but-for materiality and specific intent to deceive the USPTO, each by clear and convincing evidence. The duty-of-candor indicators in F-2 are discovery predicates, not conclusions. Note that disclosure of material prior art is not required for provisional applications.
Pre-Filing Advisory Implication
Before filing a non-provisional, counsel should assess what prior art the AI surfaced during drafting and whether any material reference requires IDS disclosure. The Background’s gap analysis language should be reviewed by counsel to ensure it accurately reflects the prior art landscape as counsel understands it, not as the AI assessed it.
AI Prior Art Forced-Choice Trap
A litigation adversary could serve the following discovery request: “Did you use any AI tool to search for or identify prior art during application preparation?”
Horn 1 — Yes
The AI prior art search results are information the applicant obtained and was aware of. If the AI surfaced specific prior art references for simulation systems, haptic feedback systems, force field generation systems, or matter replication systems that were not submitted on an IDS, that is a factual predicate for a 37 C.F.R. § 1.56 analysis as to those specific references.
Horn 2 — No
The Background section’s formal enumeration of what “conventional simulation approaches” cannot do may be difficult to explain without an AI-assisted prior art assessment. The adversary proceeds to the specification integrity argument: the AI was asked to generate a patent application and produced a Background that argues the prior art without any actual prior art search having been conducted.
Pre-Filing Advisory Implication
If the AI was used to draft the Background and the AI’s assessment of the prior art landscape is embedded in the Background language, counsel should independently assess whether the prior art identified by the AI includes any material references and whether those references require IDS disclosure before filing. Note that disclosure of material prior art is not required for provisional applications.
Authorship-Bridging Attacks Always included
AI Authorship Forced-Choice Trap
A litigation adversary could use the following cross-examination frame: “Who or what authored [specified portion of] the provisional patent application [application number] filed on [filing date]?”
Horn 1 — Attorney Authored It
The attorney filing a patent application certifies under 37 C.F.R. § 11.18 that the document’s factual contentions have evidentiary support after reasonable inquiry. Relying on AI accuracy alone does not constitute reasonable inquiry. The identified features of this application — if filed as-is — would be difficult to reconcile with meaningful attorney review. Each is an AI-generation artifact that a reviewing attorney would have modified or removed.
Horn 2 — AI Authored It
The combination of workflow artifacts (F-1), four co-occurring stylistic markers (F-3), excess claims (F-4), missing claim types (F-5), and an embedded patentability assessment (F-2) is strongly consistent with AI generation of the complete application in a single pass. If the AI authored the document, the privilege assertion over AI session transcripts is weakened, the reasonable-inquiry certification under § 11.18 is more easily challenged, and the inventorship record is incomplete.
Pre-Filing Advisory Implication
Before filing, counsel should conduct a substantive review of the document sufficient to correct all AI-generation artifacts, document the review in firm records, and establish the inventor’s conception narrative for each independent claim element. Note that the application as a whole may be jointly authored, but it can still be useful to pin down statements about authorship of particular parts of the specification or claims.
Attorney Direction Forced-Choice Trap
This is a public example document. Your actual Check A Pursuit™ report will include the full Attorney Direction Forced-Choice Trap analysis, including both horns and the pre-filing advisory implication, tailored to your specific document and proceeding type.
Privilege Stress-Test
When opposing counsel responds to AI-transcript discovery with privilege assertions, the following attack vectors are available, ranked by strength.
This is a public example document. Your actual Check A Pursuit™ report will list a set of privilege attack vectors, calibrated to the facts of your specific matter.
Scope Limitations
What This Report Covers
This report identifies conduct-based vulnerabilities arising from the behavior of the inventor, patent drafter, or patent prosecutor — specifically, indicia of AI use during drafting and the adversarial strategies those indicia support. This conduct-based analysis is within the scope of a Check A Pursuit™ engagement.
What This Report Does Not Cover
This report does not assess prior art novelty (35 U.S.C. § 102) or non-obviousness (35 U.S.C. § 103) based on an independent prior art search. It does not constitute an opinion on validity, enforceability, freedom to operate, or infringement. It does not address foreign prosecution, ownership, standing, or chain-of-title. The structural patent vulnerabilities in this draft — including the absence of CRM claims, the unbalanced claim set, and the missing drawings — are fixable drafting deficiencies that fall within Step Forward™ scope, not Check A Pursuit™ scope.
Special Limitation — Unfiled Draft
This report was generated from an unfiled draft provisional patent application. The conduct-based vulnerabilities identified above describe what a litigation adversary could do if this application were filed as-is and the resulting patent were later asserted. Several findings (particularly F-1 — workflow artifacts) are directly correctable before filing. Others (particularly F-2 — Background gap analysis language and F-3 — stylistic markers) may persist in the filed record if not corrected. This report does not assess the likelihood that any specific finding would survive a correction review by patent counsel before filing.
Completeness Disclaimer
This report reflects findings supportable from the source documents uploaded to this session. Conduct not evidenced in those materials is not addressed.
Separate Engagements
Drafting discovery requests, litigation support after the client begins acting on this report, and appearing as counsel of record are outside the scope of this engagement.