01 / Executive readingWhat the landscape indicates
The relevant prior art is not limited to “AI storytelling for education.” It also includes systems that convert activity logs into editable narratives, connect reflection with goals, construct prompts from user interactions, and link generated text to underlying sources.
Central finding
The strongest starting points are different depending on the claim you want to investigate.Read Kuato and Wright for the core narrative/self-regulation concept; ASU and 2hr Learning for AI educational storytelling; Box and Google for the technical mechanisms of activity-grounded prompting and provenance.
04Learning & storytelling precedents
04Technical implementation precedents
08Direct source links to patent records
Where patent overlap can arise
The project discussed combines curation, organisation, GenAI-assisted externalisation, and later reflection. The schematic below maps functions—not patent claims or infringement.
InputLearning tracesHighlights, annotations, saved sources
ProcessEvidence reconstructionSequence, goals, connections, uncertainties
OutputReflective narrativeStory, interpretations, linked evidence
Feedback loopLearner revisionCorrect, reflect, choose next steps
Suggested lenses: activity-to-story (No. 1), goal/reflection structure (No. 2), AI narrative personalisation (Nos. 3–4), prompt construction (No. 5), editing (No. 6), explanation (No. 7), and source attribution (No. 8).
03 / At-a-glance comparisonEight relevant patents
Ordered by relevance to the concepts and mechanisms under discussion—not by enforceability, patent quality, or infringement risk. “Overlap” is a qualitative reading, not a legal risk score.
| # | Patent record | Closest function | Relevance assessment | Important boundary |
|---|
| 01 | US20160220903A1Conceptual precedent | Activity → editable storyKuato Games (UK) Ltd. | High conceptual overlap | The published claim is specific to a virtual game, a controlled character, and an image-and-text storybook. It does not specifically claim browser annotations, programming learning, or self-regulation. |
| 02 | US8628331B1SRL mechanism | Goals ↔ reflection ↔ actionBeth Ann Wright (inventor; individual assignee listed) | High SRL-function overlap | These features are embedded in a competency-management and assessment architecture. Independent claim 1 includes structured goal selection, recorded performance, grouping, and rubric-based assessment. |
| 03 | US12272265B2AI learning narrative | Adaptive AI learning narrativeArizona Board of Regents / Arizona State University | Strong descriptive overlap | Do not equate descriptions in the specification with the scope of granted claims. The principal independent claim concerns a particular graph-data/code-generation architecture, not AI storytelling generally. |
| 04 | US20250322766A1Prompt construction | Learner-profiled story prompts2hr Learning Inc. (original applicant) | Strong prompting overlap | Centres on personalised instructional stories and reading-level adaptation, not reconstruction of the learner’s own activity history for self-regulation. Claims are published application claims, not granted claims. |
| 05 | US20240362476A1Data-to-prompt pipeline | Activities → LLM promptBox Inc. | High technical adjacency | Set within a content-management/collaboration system rather than a learning-story or self-regulation context; the application claims are not granted claims. |
| 06 | US12142298B1Editable narrative | Memory graph → editable storyMeta Platforms Inc. | Medium technical adjacency | The independent claim specifies media montages and a dialogue/memory-graph workflow. The title does not imply protection over all text-based reflective narratives. |
| 07 | US11954445B2Explanatory mechanism | Data → explanatory narrativeNarrative Science LLC (original); Salesforce Inc. (listed current) | Medium technical adjacency | Claims concern a specified structured-data/conditional narrative-generation mechanism—not merely asking an LLM to reflect on learning progress. |
| 08 | US12596889B2Evidence provenance | Generated text → source linkGoogle LLC | High provenance adjacency | Claim 1 specifies fine-tuning and a particular matching/attribution process; not every form of citation or provenance linking is covered. |
A. Learning narratives & SRL foundations
Four documents with close conceptual or educational overlap. Some patent specifications discuss storytelling more broadly than their claims protect.
01 / Narrative & learning foundations
Systems and Methods for Dynamically Creating Personalized Storybooks based on User Interactions within a Virtual Environment
Kuato Games (UK) Ltd.Priority: 2 Feb 2015Published 4 Aug 2016Published application · reported abandoned
What the relevant claim or description says
Published claim 1 records a chronological sequence of interactions in a game, associates images and descriptive text with those events, presents them as storybook pages, permits text editing, and produces a storybook file.
Relationship to your system · interpretation
An explicit precedent for converting recorded activities into a reconstructed narrative that a user can edit.
Claim-scope boundary
The published claim is specific to a virtual game, a controlled character, and an image-and-text storybook. It does not specifically claim browser annotations, programming learning, or self-regulation.
02 / Narrative & learning foundations
Learning model for competency based performance
Beth Ann Wright (inventor; individual assignee listed)Priority: 6 Apr 2010Granted 14 Jan 2014Granted patent
What the relevant claim or description says
The patent links competency goals to performance records and qualitative reflection. Claim 9 includes experiences, introspection, and next steps; claim 14 addresses sharing a story by analogy linked to goal elements.
Relationship to your system · interpretation
Strong precedent for the regulatory purpose of a reflective story: connecting experiences, goal interpretation, and subsequent action.
Claim-scope boundary
These features are embedded in a competency-management and assessment architecture. Independent claim 1 includes structured goal selection, recorded performance, grouping, and rubric-based assessment.
03 / Narrative & learning foundations
Systems, methods, and apparatuses for implementing an adaptive and scalable AI-driven personalized learning platform
Arizona Board of Regents / Arizona State UniversityPriority: 12 May 2023Granted 8 Apr 2025Granted patent
What the relevant claim or description says
The specification discusses an immersive storytelling component using language models to personalise narratives around interests, experiences, progress, achievements, and challenges. Granted claim 1 instead focuses on a learner graph database, AI-generated code to populate/update it, and queries for contextualising responses.
Relationship to your system · interpretation
Relevant to AI-generated learning narratives embedded within an adaptive learning system.
Claim-scope boundary
Do not equate descriptions in the specification with the scope of granted claims. The principal independent claim concerns a particular graph-data/code-generation architecture, not AI storytelling generally.
04 / Narrative & learning foundations
AI powered dynamic story generation system for individualized learning and a method thereof
2hr Learning Inc. (original applicant)Priority: 11 Apr 2024Published 16 Oct 2025Published application · reported pending
What the relevant claim or description says
Published claim 1 uses a learner profile, reading ability, and story topics to generate personalised stories with adjustable reading level. Claim 6 explicitly addresses prompt engineering and prompt chaining.
Relationship to your system · interpretation
Pertinent to how user context can condition AI prompts and educational story generation.
Claim-scope boundary
Centres on personalised instructional stories and reading-level adaptation, not reconstruction of the learner’s own activity history for self-regulation. Claims are published application claims, not granted claims.
05 / Patent dossiersB. Technical mechanisms
Four documents pertinent to implementation choices: activity-driven prompting, revision through dialogue, explanatory synthesis, and attribution.
05 / Technical implementation mechanisms
Generating a large language model prompt based on collaboration activities of a user
Box Inc.Priority: 30 Apr 2023Published 31 Oct 2024Published application · reported pending
What the relevant claim or description says
Published claim 1 identifies stored-content chunks, records collaboration activities tied to those chunks, and constructs an LLM prompt using their relationships. Dependent claims cover interactions such as viewing, editing, and interacting with particular content.
Relationship to your system · interpretation
Directly relevant to constructing a story prompt from annotations, resource fragments, and interaction histories.
Claim-scope boundary
Set within a content-management/collaboration system rather than a learning-story or self-regulation context; the application claims are not granted claims.
06 / Technical implementation mechanisms
Creating digital stories based on memory graphs and multi-turn dialogs for assistant systems
Meta Platforms Inc.Priority: 14 Feb 2022Granted 12 Nov 2024Granted patent
What the relevant claim or description says
Granted claim 1 creates a media montage from a user’s media collection, presents it in a dialogue, receives a change request, and produces an edited montage using that request and a user memory graph.
Relationship to your system · interpretation
Relevant to the notion of a learner-controlled evolving story grounded in linked personal records and revised conversationally.
Claim-scope boundary
The independent claim specifies media montages and a dialogue/memory-graph workflow. The title does not imply protection over all text-based reflective narratives.
07 / Technical implementation mechanisms
Applied artificial intelligence technology for narrative generation based on explanation communication goals
Narrative Science LLC (original); Salesforce Inc. (listed current)Priority: 17 Feb 2017 (family)Granted 9 Apr 2024Granted patent
What the relevant claim or description says
Granted claim 1 generates narratives from structured data using a parameterised conditional framework for selecting ideas. The narrative explains a value or change through contributing factors.
Relationship to your system · interpretation
Relevant to the difference between listing learning events and explaining meaningful changes or plausible contributing factors.
Claim-scope boundary
Claims concern a specified structured-data/conditional narrative-generation mechanism—not merely asking an LLM to reflect on learning progress.
08 / Technical implementation mechanisms
Generation of natural language (NL) based summaries using a large language model (LLM) and subsequent modification thereof for attribution
Google LLCPriority: 21 Feb 2023Granted 7 Apr 2026Granted patent
What the relevant claim or description says
Granted claim 1 fine-tunes an LLM on a third-party dataset, generates a response, compares response segments with dataset segments, and inserts a source indication and hyperlink for identified matches.
Relationship to your system · interpretation
Relevant to evidence-linked AI stories where claims can point back to a learner’s captured annotations or original source.
Claim-scope boundary
Claim 1 specifies fine-tuning and a particular matching/attribution process; not every form of citation or provenance linking is covered.
No patents match the current search. Try a broader term or switch to “All 8.”
06 / Suggested reading sequenceWhat to read first—and why
Core narrative design
Nos. 1 & 2 · Kuato + WrightInvestigate recorded events becoming editable stories, then connections between experiences, introspection, learning goals, and next steps.
AI in learning stories
Nos. 3 & 4 · ASU + 2hr LearningSeparate broad descriptions of AI-driven personalisation from the specific claims around graph systems, profiles, reading levels, and prompt chaining.
Activity-to-prompt architecture
No. 5 · BoxExamine how content chunks and recorded interactions are linked and selected for prompts; compare this carefully to annotation-grounded story generation.
Verifiable generated text
No. 8 · GoogleStudy the claimed attribution mechanism and distinguish it from other designs for citing learner notes or resources in an AI story.
07 / Interpret responsiblyWhat this report does—and does not—establish
Research orientation, not a patentability or freedom-to-operate opinion.Patent descriptions may mention much more than is claimed. Published applications may change or never issue. A granted patent’s actual scope depends on its claims, prosecution history, jurisdiction, and legal status.
- Relevance is not infringement. Similar wording or features does not demonstrate that your implementation falls within a valid patent claim.
- Prior art is broader than granted patents. Published applications, papers, demonstrations, and your own earlier disclosure may affect novelty and obviousness.
- Status needs rechecking. “Reported pending/abandoned” labels reproduce earlier retrieved records; live prosecution status and ownership were not independently verified during this formatting task.
- Claim-by-claim assessment is essential. A registered patent practitioner should review candidate claims and priority/ownership issues before any filing or launch decision.
The original research and the present report concern storytelling-centred support for informal programming learning. The eight records form a targeted orientation set, not an exhaustive patent search.
08 / Primary source linksPatent records and legal reading
- US20160220903A1 — Systems and Methods for Dynamically Creating Personalized Storybooks based on User Interactions within a Virtual Environment
- US8628331B1 — Learning model for competency based performance
- US12272265B2 — Systems, methods, and apparatuses for implementing an adaptive and scalable AI-driven personalized learning platform
- US20250322766A1 — AI powered dynamic story generation system for individualized learning and a method thereof
- US20240362476A1 — Generating a large language model prompt based on collaboration activities of a user
- US12142298B1 — Creating digital stories based on memory graphs and multi-turn dialogs for assistant systems
- US11954445B2 — Applied artificial intelligence technology for narrative generation based on explanation communication goals
- US12596889B2 — Generation of natural language (NL) based summaries using a large language model (LLM) and subsequent modification thereof for attribution
- USPTO Manual of Patent Examining Procedure — Patent Eligibility (MPEP §2103) — general legal framework for eligible patent subject matter.
Record details and claim descriptions were carried forward from the earlier analysis. The external links are embedded for verification against the official claim documents and current registers.
أضع هذا كتوثيق لتطور الموضوع و الفكرة -- سامي
ردحذف