Issuu

Enhancing Public Legal Literacy through AI-Based Storytelling and Explainable Systems

Vaishnavi

Gayakwad1, Shreya Ugemuge2, Vikas Gawade3 , Vaibhav Wadekar4 , Prof. F. I. Khandwani 5

1Final Year Student, Department of Information Technology, Shri Sant Gajanan Maharaj College of Engineering, Shegaon, Maharashtra, India

2 Final Year Student, Department of Information Technology, Shri Sant Gajanan Maharaj College of Engineering, Shegaon, Maharashtra, India

3 Final Year Student, Department of Information Technology, Shri Sant Gajanan Maharaj College of Engineering, Shegaon, Maharashtra, India

4 Final Year Student, Department of Information Technology, Shri Sant Gajanan Maharaj College of Engineering, Shegaon, Maharashtra, India

5 Shri Sant Gajanan Maharaj College of Engineering, Shegaon

Abstract - An Understanding legal information remain difficult for general public due to complex terminology and expert-oriented legal systems. Recent advancement in artificial intelligence (AI) offer a new opportunities to address this challenge by transforming abstract legal rules into intuitive narrative waste explanation. In particular, AIdriven storytelling hasthepotentialtopresentlegalconcepts through the relatable scenarios that align with everyday experience.

This paper reviews existing research in a computational storytelling, AI-based learning systems, legaltechnology, and explainable artificial intelligencetoexaminetheirrelevanceto the public legal education. Basedonthisreview, anAI-driven storytelling framework is proposed that generates accurate and personalized narratives grounded in the structured legal data. The framework adapts explanations to user context and information needs, enablinglegalrightsandresponsibilitiesto be communicated in a clear and accessible manner.

By combining legal knowledge grounding with adaptive storytelling and explainable AI techniques, the proposed approach aims to reduce barriers to legal understanding for non-expert user, including tenants, that small business owner and the other everyday citizens. This work highlight the potential of AI enable narratives as a scalable solution for improving public legal literacy.

1. INTRODUCTION

Legalawarenessisessential forempoweringindividual to understand their rights, duties, and available courses of action. However, legal knowledge is often highly complex technical,andprimarilydesignforexperts,makingitlargely inaccessibletothegeneralpublic.Despitegreateraccessto the legal resources via internet, a significant gap remains betweenavailableknowledgeandpubliccomprehension. Recent advancement in artificial intelligence (AI) have demonstrateasignificantpotentialinaddressingchallenges

related to the information accessibility and learning. AIbased educational systems support personalized learning, adaptive content delivery, and automated knowledge generation, thereby improving user engagement and comprehension across a diverse domains. In parallel, computational storytelling has emerged as an effective approachforpresentingcomplexconceptthroughnarrative structures, allowing users to contextualize abstract informationwithinrelatablereal-worldscenarios.

However, the application of AI-driven storytelling techniquesinalegaldomainremainslimited.Mostexisting storytelling and generative AI systems operate without domain-specific constraints, raising concerns related to factual accuracy, interpretational validity and ethical reliability.Theselimitationsare particularlycriticalinlegal contexts,whereinaccurateormisleadingexplanationscan resultinaseriousconsequenceforauser.

Conversely,AIapplicationsdevelopedspecificallyforalegal womanhaveprimarilyfocusesonanalyticalandpredictive tasks, such as legal document classification, case outcome prediction, and legal research assistance. While these systemsdemonstrateastrongperformanceinprofessional legal environments, they are typically designed for use by legal experts. The explanations they provide are often technical and model-centric offering limited support for communicating legal concepts in an intuitive and user friendlymannertoageneralpublic.

Explainable Artificial Intelligence is essential for building transparency and trust in AI systems, particularly in sensitivedomainssuchaslawandeducation.However,most existing Explainable AI methods focus on technical explanations that are difficult for non-expert users to understand and offer limited support for learning. This highlights the need for an approach that combines legal knowledge with narrative-based, personalized, and explainable explanations. Accordingly, this paper reviews

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 13 Issue: 02 | Feb 2026 www.irjet.net p-ISSN: 2395-0072

existing research and proposes an AI-driven legal storytellingframeworktoenhancepubliclegalliteracy.

2. LITERATURE SURVEY

Artificial intelligence (AI) is increasingly being used in learning support systems, especially in fields that involve complex and specialized knowledge. Research spanning educational technologies, digital storytelling, and legal informatics indicates that traditional text-centric system have often inadequate or non-expert users, as they fail to conveyabstractortechnicalconceptinanintuitivemanner. Thelimitationshasmotivatedtheexplorationofalternative approaches that combines AI with a narrative-driven and explainablemechanismtoimproveclarityandaccessibility forageneralusers.

Computational storytelling research focuses on the automated creation, organization, and adaptation of narrativesusingalgorithmictechnique.Studiesinthisarea demonstrate that digital narratives need not follow a fix linearstructure,insteadtheycanevolvedynamicallybased onthesystemlogicoruserinteraction.Suchadaptiveand emergent storytelling models emphasis on narrative coherence,userengagementandflexibility.Howevermost computationalstorytellingapplicationsremainconfidentto entertainmentdomainsuchasgames,films,museums,and creativemedia,withalimitedapplicationineducationalor legalknowledgedimensionalcontext[1].

Recent advances in large-scale generative AI models have futureenhancedthecapabilityofsystemstoproducefluent, human-like narratives and explanation with the minimal human intervention. While this model shows promise in automatingcontentgeneration,existingliteraturehighlights critical concern related to the factual correctness, controllability, and ethical reliability. These concerns becameparticularlysignificantinhigh-staketowerssuchas law, where inaccurate or biased explanation can lead to a seriousconsequence[4],[5].

Withintheeducationaldomain,AI-drivensystemshavebeen extensively studied for personalization and adaptive learning.Priorresearchdemonstratesthatlearnermodelling anddynamicsequencingofcontentcansignificantlyimprove learner engagement and comprehension. However, most existing educational AI system prioritize instructional efficiencyandassessmentperformance,oftenoverlooking narrative-basedorexperimentalexplainablemethods.Asa result,learnersmaystruggletoemotionallyconnectwithor contextualizeinformationpresentedtothem[3],[7].

Legal AI research has largely focused on automating analytical tasks such as legal document classification, judgment prediction, and case outcome analysis. The development of a large-scale legal datasets, particularly thosetailoredtotheIndianjudicialsystem,hasimprovedthe

performanceofsuchsystemandhighlightedtheimportance of an explanationin legal decision- making. Despite these advancements,currentsystemsareprimarilydesignedfora legalprofessionalsandofferalimitedsupportforsimplifying alegalconceptsforageneralpublic[6].ExplainableAIhas emergedasacriticalrequirementforAIsystemsoperating inasensitivedomainssuchaseducationandlaw.Existing explainable AI approaches aim to make model decisions transparentbutoftenrelyontechnicalexplanationsthatare difficultforanon-exportuserstointerpret.Narrative-based explanations, although potentially more intuitive userfriendly,remainunexploredwithinalegalAIresearch[9].

Furthermore. Studies in a cognitive science and learning psychology suggest that storytelling enhances comprehensionandretentionbytheembeddinginformation within the relatable and emotionally engaging contexts. These finding indicate that narrative-based explanations may be particularly effective for conveying complex legal concepts in an accessible and meaningful manner In summary, while significant progress has been achieved independently in computational storytelling, AI-driven education, and legal natural language processing, and explainable AI, existing literature reveals a clear lack of unifiedsystemsthatintegratelegalknowledgegrounding, narrativegeneration,personalization,andexplainabilityto promotepubliclegalawarenessandunderstanding.

3. COMPARATIVE STUDT AND ANALYSIS

To identify key limitations in existing approaches, a comparative analysis of computational storytelling, AIdriveneducation,legalAI,andexplainableAIisconducted. The comparison is based on four dimensions: domain focus, explanation style,adaptabilitytousers,andtarget audience. This analysis highlights how current systems emphasizecertainstrengthswhileoverlookingothers.

Table -1: Comparativeanalysisofexistingapproaches

Approac h Storyte lling Personali zation Explaina bility Target Users

Computa tional Storytelli ng Strong Limited Narrative coherenc e Entertain ment users

AI-based Educatio nal Systems Minimal Strong Performa nceoriented Students and educator s

Generativ e AI Content Implicit Generic Limited factual control General users

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Systems

Legal AI and NLP Systems

Explaina ble AI (XAI)

Absent Low Technical , modelcentric Legal professio nals

Rare Low Technical transpare ncy AI experts

This comparative analysis indicates that computational storytellingsystemsarehighlyeffective in engagingusers throughadaptiveandcoherentnarratives.However,these system generally lack grounding in a factual and domainspecificknowledge,particularlyinlegalsensitivecontexts.In contrast,legalAIsystemsdemonstrate stronganalyticaland data-drivencapabilitiesbutoftenprovideexplanationsthat are technical and difficult for a non-export user to comprehend.

AI-driven educational systems effectively personalize content based on the learner characteristics, improving engagementandcompression.Nevertheless,thesesystems rarely employ narrative-based explanations, limiting contextual understanding and emotional connection for learners. Each of these approaches address of specific dimension-storytelling, legal reasoning or personalization whileneglectingothers.

GenerativeAImodelspartiallybridgethesegapsbyenabling automatednarrativegeneration.Despitetheirpotentialthey introduce challenges related to factual reliability, controllability, and ethical fairness, which are particularly critical in the legal application. Similarly, explainable AI techniquesaimtoenhancetransparencybutoftenrelyon abstractonatechnicalexplanationsthatareinaccessibleto thegeneralusers.

Overall, existing approach tends to address storytelling, personalization, explainability, and legal reasoning in isolation. The absence of an integrate framework that combineslegallygroundedknowledge,adaptingstorytelling, user-centricpersonalization,andintuitivetoexplainability significance limit the effectiveness of a current system in promptingpubliclegalawarenessandunderstanding.

4. RESEARCH GAP AND CHALLENGES

Although notable progress has been made in a computationalstorytelling,AI-basedlearningsystems,legal artificialintelligence,andexplainableAI,existingliterature reveals several gaps that limit their effectiveness in promotingpubliclegalawareness.

A major gap lies in the absence of unified systems that combine legally grounded and validated knowledge with

narrative-based explanations. Computational storytelling approachesprioritizeengagementandnarrativecoherence but typically operate without legal constraints or domain validation,makingthemunsuitableforexplainingreal-world legal concepts that demand accuracy and interpretational correctness.Incontrast,legalAIsystemsfocusonanalytical and predictive tasks and are primarily designed for legal professionals, offering limited support for communication legalinformationinanaccessiblemannertonon-expert user.

AI-based educational platforms effectively address personalization and adaptive learning through structured content delivery. However, they generally rely on instructional formats rather than narrative-driven explanations.Story-basedlearning,whichfostersemotional engagement and contextual understanding, is particularly important for explaining abstract and interpretative domainssuchaslaw,yetremainsinsufficientlyexploredin current educational AI systems. Generative AI models enable efficient narrative generation but introduce challengesrelatedtofactualreliability,controllability,and ethical alignment, especially in legally sensitive contexts. Similarly, existing XAI techniques aim to improve transparency but often emphasize model-centric explanations, such as feature importance scores or rulebased outputs. While useful for exports this explanation formats are difficult to lay users to interpret and do not adequatelysupportuserscomprehension.

In addition to these conceptual gaps, several practical challengespersist.Balancinglegalprecisionwithanarrative simplicity is inherently complex, particularly when explanations must remain accurate yet understandable. Ensuring adaptability across users with a diverse backgroundsandliteracylevelsfurthercomplicatessystem design. Moreover, though limited availability of a highquality publically accessible by legal datasets especially withintheIndianlegalcontextposesasignificantchallenge toscalableandreliableimplementation.

Overall, the literature highlights a clear need for an integrated framework that combines legally grounded knowledge,adaptivestorytelling,personalization,andusercentricexplainability.Addressingthesegapsisessentialfor developingAIsystemsthatnotonlyanalyzelegaldatabut alsomakelegalknowledgeaccessible,understandable,and meaningfulforgeneralpublic.

5. PROPOSED FRAMEWORK

Toaddresstheidentifiedresearchgaps,thispaperpurposes a holistic AI-driven framework that integrates legally groundedknowledgerepresentation,adaptivestorytelling, presonalization,andexplainabilitytoenhancepubliclegal awareness. The proposed framework aims to explain complex legal provisions through simple, relatable

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 13 Issue: 02 | Feb 2026 www.irjet.net p-ISSN: 2395-0072

narratives while ensuring factual accuracy, transparency, andethicalcompliance.

Framework Overview

The proposed system follows a modular architecture in whichmultipleinterconnectedcomponentscollaboratively transform legal information into user-friendly narrative explanations. The framework is designed to balance legal correctnesswithusability,ensuringthatgeneratedcontent remains both legally sound and easily understandable for non-expertusers.

Key Components of the Framework

Legal Knowledge Repository

Thiscomponentservesasthefoundationalknowledgebase of the system. It comprises structured legal documents, statutes, case summaries, and judicial interpretations extractedfromtrustedlegalcorpora,includingIndianlegal datasets such as Indian Legal Documents Corpus (ILDC). Legal content is preprocessed, annotated, and indexed to ensure traceability and consistency during narrative generation[6]

Legal Concept Extraction and Mapping Module

This module indentifies relevant legal concepts, entities, rights,duties,andpenaltiesfromtheknowledgerepository. Extractedconceptsaremappedtoreal-worldcontextsand everydayscenarios,enablingabstractlegalprovisionstobe transformedintomeaningfulnarrativeelementssuitablefor storytelling[1].

Narrative Generation Engine

The narrative generation engine convert mapped legal conceptsintostructuredstories.Unlikeunconstrainedstory generation, this module relies on predefined narrative templates and legal constraints to preserve factual correctness.Storyprogressionisgovernedbycausallogic, ensuring that each narrative event reflects valid legal reasoningratherthanimaginativeimprovisation[1],[4].

Personalization and Iser Modeling Model

To enhance user engagement and comprehension, the framework incorporates user modeling techniques that dynamicallyadjustnarrativecomplexity,languagelevel,and contextual examples. Personalization is based on factors suchastheuser’slegalliteracy,educationalbackground,and interactionhistory,aligningwithprinciples observedinAIbasededucationalsystems[3],[7].

Explainable Storytelling Layer

TheexplainableAIcomponentembedsexplainationsdirectly within the narrative flow. Legal reasoning, consequences, anddecisionpointsareclarifiedusingintuitivestorytelling elements rather than technical model outputs. This narrative-based explainability bridges the gap between modeltransparencyanduserunderstandability[9].

Validation and Ethical Compliance Module

Thismoduleverifiesgeneratednarrativesagainstlegalrules andethicalconstraintstopreventmisinformation, bias,or hallucinated content. It ensures compliance with a legal standards and address challenges associated with the generativeAIsystemsinasensitivelegaldomains[4],[5].

User Interaction and Feedback Module

Thesystemsupportsinteractiveuserengagement,allowing users to ask follow-up questions, seek clarifications, or explorerelatedlegalconcepts.Userfeedbackiscontinuously collected and utilized to improve narrative quality, personalizationaccuracy,andoverallsystemreliability.

Framework Workflow

Theworkflowbeginswithuserinput,suchasalegalquery orselectedtopic.Relevantlegalknowledgeisretrievedfrom therepositoryandprocessedforconceptextraction.These concepts are then mapped into a legally constrained narrative, which is personalized according to user characteristicsandenrichedwithembeddedexplanations. Thegeneratedstoryundergoesvalidationandethicalchecks beforebeingdeliveredtotheuser.Feedbackcollectedduring interactionisusedforiterativesystemrefinement.

Significance of the Proposed Framework

By integrating legally grounded knowledge, adaptive storytelling, personalisation, and explainable AI, the proposedframeworkovercomesthelimitationsofexisting fragmented approaches. The framework offers a scalable solution for delivering legal information that is not only accuratebut alsoaccessible, engaging, and meaningful for thegeneralpublic,therebysupportingbroaderlegalliteracy andinformedcivicparticipation.

6. FUTURE SCOPE

TheproposedAI-enabledlegalstorytellingframeworkoffers several promising directionsforfutureenhancement. One importantextensionistheinclusionofmultilingualsupport, whichimproveaccessibilityforusersfromdiverselinguistic backgrounds,particularlyintheIndiancontext.Inaddition, the framework can be future strengthened by supporting multimodal legal storytelling through the integration of audio and visual elements. This would help accommodate userwithvaringliteracylevelsandlearningpreferences.

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 13 Issue: 02 | Feb 2026 www.irjet.net p-ISSN: 2395-0072

Futureworkmayalsofocusonadvancedpersonalizationby modelinglongtermuserbehavioranddevelopingadaptive learningpaathways.Suchanapporachwouldenableusersto gradually bulid legal knowledege over time, rather than receiving isolated explanations. Futhermore, integrating real-timeconnectionstoofficallegaldatabaseswouldallow thesystemtostayupdated withrecentlegal amendments andjudicialdevelopments.Finallylarge-sacleuserstudies and robust evaluation mechanisms can be conducted to assessusercomprehension,fairnessandoverallusabilityof AI-basedlegalexplanationsystems.

7. CONCLUSION

This paper analysed computatioanl storytelling, AI in education, legal AI, and explainable AI to examine their strengths and limitations in enhancing legal awareness amongthepublic.CurrentapproachesAddressengagement, personalization,andlegalaccuracyseparately,butnotinthe integratedmannerforimprovingpublicunderstandingofa law. To bridge this gap, an AI-assistant legal storytelling frameworkhasbeenproposed,combininglegallygrounded knowledge, narrative generation, personalization, and explainablestorytelling.Theframeworkaimstoprovidethe accessible, engaging, and credible legal information to lay people. Future enhancement could include multilingual support,multimodalstorytellingreal-timeupdatesofalegal information.Overall,thisapproachoffersscalablesolutionto narrow the divide between complex legal knowledge and publiccompression,leveragingartificialintelligencetomake legalstorytellingeffectiveandwidelyaccessible.

REFERENCES

[1] G. Trichopoulos, G. Alexandridis, and G. Caridakis, “A SurveyonComputationalandEmergentDigitalStorytelling,” Heritage,vol.6,no.2,pp.1227–1263,2023.

[2] R. Gu, H. Li, C. Su, and W. Wu, “Innovative Digital StorytellingwithAIGC:ExplorationandDiscussionofRecent Advances,” arXiv preprint arXiv:2309.14329,2023.

[3] B. Zawacki-Richter, et al., “AI Applications in Higher Education:ASystematicReview,” Int. J. Educ. Technol. High Educ.,vol.16,pp.1–34,2019.

[4] Y. Wang, J. Lin, Z. Yu, W. Hu, and B. F. Karlsson, “Open-WorldStoryGenerationwithStructuredKnowledge Enhancement: A Comprehensive Survey,” arXiv preprint arXiv:2212.04634,2022.

[5]N.Malik,R.Sanjay,S.K.Nigam,andK.Ghosh,“ILDCfor CJPE: Indian Legal Documents Corpus for Court Judgment PredictionandExplanation,” Proc. 59thAnnu. Meet. Assoc. for Computational Linguistics & 11th Int. Joint Conf. on Natural Language Processing (Volume 1: Long Papers), pp. 4046–4062,2021.

[6] T. B. Brown, et al., “Language Models are Few-Shot Learners,”in Adv. Neural Inf. Process. Syst.,vol.33,pp.1877–1901,2020. (Generative models reference)

[7] “AI-Based Learning Content Generation and Pathway AugmentationtoIncreaseLearnerEngagement,” Computers & Education: Artificial Intelligence

[8] Y. Mori, H. Yamane, Y. Mukuta, and T. Harada, “ComputationalStorytellingandEmotions:ASurvey,” arXiv preprint arXiv:2205.10967, 2022. (Story and emotion connection reference)

[9] A. Barredo Arrieta, N. Díaz-Rodríguez, and others, “Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward ResponsibleAI

[10]H.Johnston,R.F.Wells,E.M.Shanks,T.Boey,andB.N. Parsons, “Student Perspectives on the Use of Generative Artificial Intelligence Technologies in Higher Education,” International Journal for Educational Integrity,vol.20,Art.2, 2024