{"id":11290,"date":"2026-03-11T15:41:58","date_gmt":"2026-03-11T15:41:58","guid":{"rendered":"https:\/\/ipauthor.com\/?p=11290"},"modified":"2026-07-29T19:55:49","modified_gmt":"2026-07-29T14:25:49","slug":"the-next-phase-of-patent-ai-agentic-prosecution-workflows-chemistry-support-and-smarter-office-action-analysis","status":"publish","type":"post","link":"https:\/\/ipauthor.com\/?p=11290","title":{"rendered":"The Next Phase of Patent AI:
Agentic Prosecution Workflows, Chemistry Support, and Smarter Office Action Analysis"},"content":{"rendered":"\t\t
\n\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

A patent’s commercial value is rarely decided in the initial drafting. It is won or lost in the Office Action response. That is where prosecution history estoppel is born, and claim scope is either surgically defended or inadvertently signed away.<\/p>\n

For most prosecution teams, the bottleneck is not the legal reasoning. It is the structural work that has to happen before the reasoning can even begin. Confirming the right claim language, reconstructing the examiner’s mapping logic, verifying cited references, and maintaining consistency across every section of the response. IP Author’s latest release addresses each of those friction points directly.\u00a0<\/p>\n

Four updates, now live in IP Author, each targeting a specific place where the current workflow slows attorneys down or introduces unnecessary risk.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t\t

1. Claim Editor: Lock the Foundation Before Arguments Begin<\/h2>\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

In any Office Action response, the claims are the starting point. If you are working from the wrong version, every argument you build on top of that is at risk. And this is easier to miss than it sounds, especially after multiple amendment rounds or when more than one person is working on the same matter.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t

\n\t\t\t\t
\n\t\t\t\t\t

The Problem This Solves<\/h3>\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\tVersion drift is real. After several rounds of amendments, attorneys can build an entire response strategy on a claim set they assumed was current but was not. The misalignment usually surfaces at the end, when rework is expensive and the deadline is close.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

IP Author now requires attorneys to confirm the current claim set before any drafting begins. The Claim Editor puts the claims in front of you for review. You lock them. That confirmed version becomes the foundation for every argument, every amendment, and every strategic decision that follows. Nothing downstream gets built on an assumption.<\/p>\n

There is also a practical escape valve. The USPTO-processed version of your claims does not always match what you have on file. It is not common, but it happens. If it does, you can upload your own version directly as a PDF or Word document. The platform reads it and treats it as the confirmed baseline going forward.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t\t

What This Means for Your Practice<\/h3>\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t\t\t\t
    \n\t\t\t\t\t\t\t
  • \n\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t<\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\tEvery argument gets built on claims you have personally reviewed and confirmed<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t
  • \n\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t<\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\tNo version surprises discovered at the end of a response cycle<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t
  • \n\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t<\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\tUpload support for PDF and DOCX when the USPTO version needs to be overridden<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t
  • \n\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t<\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\tConsistent claim foundation across multi-attorney matters where version discipline is hardest to enforce<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t<\/ul>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t
    \n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\"\"\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t
    \n\t\t\t\t\t

    2. Chemistry Office Action Support: One Platform for Every Technical Discipline<\/h2>\t\t\t\t<\/div>\n\t\t\t\t
    \n\t\t\t\t\t\t\t\t\t

    Chemistry prosecution has always been the odd case out. Pharmaceutical and speciality chemical matters required separate handling, independent tracking, and a different analytical approach. Teams have worked around this for years by routing these cases outside the standard workflow.<\/p>\n

    That workaround is no longer necessary.<\/p>\n

    IP Author now supports Chemistry Office Actions within the same structured workflow used for software, mechanical, and biomedical cases. For prosecution teams handling pharmaceutical or speciality chemical matters, this means everything runs through one platform with one record.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t

    \n\t\t\t\t
    \n\t\t\t\t\t

    Why Chemistry Is Different and Why That Matters<\/h3>\t\t\t\t<\/div>\n\t\t\t\t
    \n\t\t\t\t\t\t\t\t\tChemistry prosecution does not work on text matching. Claim scope turns on structural similarity. Examiners cite SciFinder, CAS, and Reaxys rather than relying only on patents and journals. Obviousness rejections focus on whether a skilled chemist would have modified a specific substituent on a known scaffold. Novelty rejections may rely on structural analogs that were never expressly disclosed in text. Genus and species distinctions, Markush claim scope, overlapping ranges, unexpected results, and inherency are all routine. IP Author now handles all of it within the standard workflow.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
    \n\t\t\t\t\t

    What the Platform Now Does<\/h3>\t\t\t\t<\/div>\n\t\t\t\t
    \n\t\t\t\t\t\t\t
      \n\t\t\t\t\t\t\t
    • \n\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t<\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\tProcesses chemistry-specific claim language including Markush groups and genus\/species distinctions<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t
    • \n\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t<\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\tMaps examiner objections based on structural analogs directly to the claimed structure and functional limitations<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t
    • \n\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t<\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\tEvaluates database-based novelty and obviousness rejections as structural comparisons, not simple text matches<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t
    • \n\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t<\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\tRecognises SMILES, MOL, and SDF chemical structure formats uploaded directly, displayed as-is without distortion<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t
    • \n\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t<\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\tIdentifies core scaffolds, substitution positions, and structural features relevant to claim differentiation<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t
    • \n\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t<\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\tSupports Markush-style claim drafting with full platform integration<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t<\/ul>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t
      \n\t\t\t\t\t\t\t\t\tThere is also a secondary benefit that extends beyond chemistry matters. The image analysis capability built to support chemical structures now lets the platform visually analyse figures in any patent application. For mechanical devices and biomedical technologies, the system reads images as images. It maps what is depicted and works those visual cues into the strategy it recommends. For complex mechanical or biomedical cases, this is going to make a noticeable difference in the precision of the analysis.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t
      \n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\"\"\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t
      \n\t\t\t\t\t

      3. European Office Action Analysis: Structured Mapping, Single Workspace<\/h2>\t\t\t\t<\/div>\n\t\t\t\t
      \n\t\t\t\t\t\t\t\t\t

      EPO examination is precise by design. Every objection is built around a specific document, a specific passage, and a specific claim element. Getting to a response requires knowing exactly which claims are being challenged, which passages the examiner relied on, and where the argument needs to be constructed.<\/p>\n

      In practice, that mapping work has always fallen to the attorney. Pulling the examiner’s narrative apart, rebuilding it as a structured picture of which reference addresses which claim, locating the relevant passages. It is careful and necessary work. It is also exactly the kind of structural groundwork the platform can take on.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t

      \n\t\t\t\t
      \n\t\t\t\t\t

      The Structural Groundwork Is Now Handled Before You Begin<\/h3>\t\t\t\t<\/div>\n\t\t\t\t
      \n\t\t\t\t\t\t\t\t\tThe enhanced European Office Action workflow surfaces the examiner’s full citation-to-claim structure before analysis begins. Novelty objections and inventive step challenges are presented with clear visibility into which reference is cited, which passage is being relied on, and which claim element is at issue. Attorneys move directly from understanding the rejection to evaluating the response path, without having to first reconstruct what the examiner actually said.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
      \n\t\t\t\t\t

      What This Changes in Practice<\/h3>\t\t\t\t<\/div>\n\t\t\t\t
      \n\t\t\t\t\t\t\t
        \n\t\t\t\t\t\t\t
      • \n\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t<\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\tNovelty objections are mapped to specific claims and passages automatically, without manual reconstruction<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t
      • \n\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t<\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\tInventive step challenges are presented with clear citation-to-claim traceability before any strategy work begins<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t
      • \n\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t<\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\tThe legal analysis stays with the attorney; the structural groundwork moves to the platform<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t
      • \n\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t<\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\tAll analysis runs within a single workspace, so there is no need to cross-reference between the examiner's letter, the cited documents, and separate notes<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t<\/ul>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t
        \n\t\t\t\t\t

        4. Universal Chatbot in the US Office Action Workflow: Context-Aware Assistance Without Leaving the Platform<\/h2>\t\t\t\t<\/div>\n\t\t\t\t
        \n\t\t\t\t\t\t\t\t\t

        US Office Action practice requires continuous evaluation. You are not just interpreting the examiner’s reasoning. You are also pressure-testing claim language, checking specification support, exploring amendment paths, assessing fallback positions, and thinking through how each response decision will shape the prosecution record downstream.<\/p>\n

        That analytical process generates questions at every stage. And until now, finding the answers meant leaving the platform.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t

        \n\t\t\t\t
        \n\t\t\t\t\t

        The Agentic Layer: What It Actually Does<\/h3>\t\t\t\t<\/div>\n\t\t\t\t
        \n\t\t\t\t\t\t\t\t\tThe Universal Chatbot is integrated directly into the US Office Action workflow. It has context-aware access to the entire matter: the confirmed claim set, the examiner’s rejection, the cited references, every prior Office Action, and every prior applicant response. It does not just answer questions. It navigates between screens, generates strategies, validates drafts, and surfaces specific passages the examiner relied on. You never have to lose the thread of what you are working on.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
        \n\t\t\t\t\t

        What You Can Do Inline<\/h3>\t\t\t\t<\/div>\n\t\t\t\t
        \n\t\t\t\t\t\t\t
          \n\t\t\t\t\t\t\t
        • \n\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t<\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\tAsk why the examiner cited a particular reference and identify exactly which passage they relied on<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t
        • \n\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t<\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\tRequest a structural comparison between the examiner's cited compound and your claimed structure<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t
        • \n\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t<\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\tExplore how a proposed amendment would affect overall claim scope and prosecution history<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t
        • \n\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t<\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\tValidate specification support for any claim limitation without leaving the drafting environment<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t
        • \n\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t<\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\tCheck antecedent basis, claim structure, and whether key aspects of the invention are fully captured<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t
        • \n\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t<\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\tPaste in your own draft response and ask the system to check it against the examiner's position<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t
        • \n\t\t\t\t\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t<\/path><\/svg>\t\t\t\t\t\t<\/span>\n\t\t\t\t\t\t\t\t\t\tAsk about statutory standards, fallback positions, and how a proposed revision may shape the record going forward<\/span>\n\t\t\t\t\t\t\t\t\t<\/li>\n\t\t\t\t\t\t<\/ul>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t
          \n\t\t\t\t\t\t\t\t\tThe system also incorporates the full prosecution history of the matter. Every prior rejection and every prior applicant response is in scope when strategy recommendations are generated. This is not a chatbot answering questions in isolation. It is an agentic assistant with continuity across the full life of the case.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t
          \n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\"\"\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t
          \n\t\t\t\t\t

          A Prosecution Scenario: How These Four Updates Work Together<\/h2>\t\t\t\t<\/div>\n\t\t\t\t
          \n\t\t\t\t\t\t\t\t\tA pharmaceutical company has a patent application for a novel drug compound. The examiner pushes back with two rejections: the compound is structurally similar to a known reference, and two prior publications combined, the examiner argues, make it obvious. Here is how IP Author supports the attorney’s response, from the first step to the final argument.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t
          \n\t\t\t\t\t

          Step 1: Lock the Claims (Claim Editor)<\/h3>\t\t\t\t<\/div>\n\t\t\t\t
          \n\t\t\t\t\t\t\t\t\tBefore any analysis begins, IP Author asks the attorney to confirm the current claim set. After multiple amendment rounds, it is easy to assume the claims are exactly as you remember them. Sometimes they are not. The attorney reviews the claims, confirms them, and locks them in. Everything that follows is built on that verified foundation.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t
          \n\t\t\t\t\t

          Step 2: Clarify the Examiner's Position (Universal Chatbot)<\/h3>\t\t\t\t<\/div>\n\t\t\t\t
          \n\t\t\t\t\t\t\t\t\tThe examiner references a database compound and argues structural similarity. The attorney is not certain exactly which part of the database entry the examiner is relying on. Rather than leaving the platform to investigate, they ask the chatbot directly. It identifies the specific passage and notes that the comparison rests on implied structural similarity rather than an explicit structural match. That distinction becomes the starting point for the novelty argument.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t
          \n\t\t\t\t\t

          Step 3: Build the Chemistry Response (Chemistry Support)<\/h3>\t\t\t\t<\/div>\n\t\t\t\t
          \n\t\t\t\t\t\t\t\t\t

          The platform maps the examiner’s cited compound against the claimed structure and highlights the structural differences that matter. It also surfaces the functional limitation in the confirmed claims, specifically the performance threshold the drug must meet, that the examiner’s obviousness analysis does not account for. The attorney can see clearly where the examiner’s argument falls short and build the response around those distinctions, all within the same workflow.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t

          \n\t\t\t\t
          \n\t\t\t\t\t

          The Result<\/h3>\t\t\t\t<\/div>\n\t\t\t\t
          \n\t\t\t\t\t\t\t\t\tThe attorney submits a response that directly challenges the structural similarity argument and addresses the obviousness rejection with a clear, claim-anchored rationale. No switching between tools. No rebuilding context in the middle of drafting. No reconciliation pass at the end. The claim foundation was confirmed at the start, the examiner’s position was clarified inline, and the chemistry analysis was structured before the first argument was written.\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
          \n\t\t\t\t\t

          Before and After: At a Glance<\/h2>\t\t\t\t<\/div>\n\t\t\t\t
          \n\t\t\t\t\t\r\n