THE ROLE OF GENERATIVE AI IN THE AUTOMATION OF LEGAL DOCUMENT PREPARATION: AN ANALYSIS OF ACCURACY, EFFICIENCY, AND ETHICAL ASPECTS

Keywords: generative artificial intelligence, large language models, legal document automation, Retrieval-Augmented Generation (RAG), AI accuracy, legal technology, AI ethics, data verification

Abstract

This article explores the transformative impact of generative artificial intelligence (AI) on legal practice, with a particular focus on the preparation and analysis of legal documents. It examines modern generative AI models, their capabilities, limitations, and the risks associated with their implementation in the legal domain. The aim of the article is to analyze the current state of generative AI applications for automating legal tasks, identify key issues related to the accuracy and reliability of generated content, and propose a hypothetical model to enhance the trustworthiness of AI-generated outcomes. To achieve this, the article sets out the following objectives: to review recent academic research and publications evaluating the effectiveness and precision of large language models (LLMs) in legal tasks; to identify the unresolved issue of factual verification of generated content; and to formulate and justify a hypothesis for improving baseline research results through the implementation of a hybrid architecture. The study employs methods of systems analysis, comparative analysis of empirical studies, synthesis, and modeling. It analyzes foundational research on structured approaches to AI-assisted legal document analysis and the outcomes of comparative tests between LLMs and human legal professionals, highlighting the persistent challenge of factual accuracy in AI-generated content The article proposes a hypothesis that integrating structured prompting techniques with a Retrieval-Augmented Generation (RAG) architecture-leveraging a curated and dynamically updated legal knowledge basecan significantly improve the factual precision and reliability of AI-generated legal documents. This hypothesis is substantiated with a formalized expression and a potential mechanism for implementation and verification. The proposed approach is shown to not only minimize the risk of AI "hallucinations" but also to lay the groundwork for the development of more responsible and ethical tools that augment, rather than replace, the professional competencies of legal practitioners.

References

1. McCallon, J. Legal Tech ROI: Measuring AI Implementation Success. CallidusAI Blog. 2025. URL: https://callidusai.com/legal-tech-roi-measuring-ai-implementation-success/ (дата звернення: 22.06.2025)
2. Terzidou K. Generative AI systems in legal practice: Offering quality legal services while upholding legal ethics. International Journal of Law in Context. 2025.
URL: https://www.cambridge.org/core/journals/international-journal-of-law-in-context/article/generative-ai-systems-in-legal-practice-offering-quality-legal-services-while-upholding-legal-ethics/34011A84AA58A2BAB556A406A4653A8D/ (дата звернення: 22.06.2025)
3. Braff D. AI is the future of law, but most legal pros aren't trained for it, a new report says. ABA Journal. 2025. URL: https://www.abajournal.com/web/article/ai-is-the-future-of-law-but-most-legal-pros-arent-trained-for-it-a-new-report-says/ (дата звернення: 22.06.2025)
4. Martin L., Whitehouse N., Yiu S., Catterson L. Perera R. Better Call GPT, Comparing Large Language Models Against Lawyers. arXiv preprint arXiv:2401.16212. 2024. URL: https://arxiv.org/abs/2401.16212/(дата звернення: 22.06.2025)
5. Baxter J. AI-driven legal tech trends for 2025. NetDocuments Blog. 2025. URL: https://www.netdocuments. com/blog/ai-driven-legal-tech-trends-for-2025/ (дата звернення: 22.06.2025)
6. Schwarcz D., Manning S., Barry P., Cleveland D., Prescott J., Rich B., Deakins O. AI-Powered Lawyering: AI Reasoning Models, Retrieval Augmented Generation, and the Future of Legal Practice. Learning Ecosystems. 2025. URL: http://danielschristian.com/learning-ecosystems/2025/05/22/ai-powered-lawyeringby-various-authors-other-legaltech-related-items/ (дата звернення: 22.06.2025)
7. Davenport M. Enhancing legal document analysis with Large Language Models: A structured approach to accuracy, context preservation, and risk mitigation. Scientific Research an Academic Publisher. 2025. URL: https://www.scirp.org/journal/paperinformation?paperid=141892/ (дата звернення: 22.06.2025)
8. Perlman A. The Legal Ethics of Generative AI. Suffolk University Law Review. 2024. URL: https://bpb-us-e1.wpmucdn.com/sites.suffolk.edu/dist/3/1172/files/2025/02/1FINAL_Perlman_Article_WDFF.pdf/ (дата звернення: 22.06.2025)
9. Donald D. Bias in AI Large Language Models: Risks and Remedies. GPSolo Magazine. 2025. URL: https://www.americanbar.org/groups/gpsolo/resources/magazine/2025-mar-apr/bias-ai-large-language-models-risks-remedies/ (дата звернення: 22.06.2025)
10. Johnston P. Retrieval-Augmented Generation (RAG): Towards a promising LLM architecture for legal work? Harvard Journal of Law & Technology: JOLT Digest. 2025. URL: https://jolt.law.harvard.edu/digest/retrieval-augmented-generation-rag-towards-a-promising-llm-architecture-for-legal-work (дата звернення: 22.06.2025)
11. Magesh V., Surani F., Dahl M., Suzgun M., Manning C., Ho D. Hallucination-Free? Assessing the Reliability of Leading AI Legal Research Tools. Journal of Empirical Legal Studies. 2025. URL: https://dho.stanford.edu/wp-content/uploads/Legal_RAG_Hallucinations.pdf/ (дата звернення: 22.06.2025)
12. Kankatala, V. Retrieval-Augmented Generation (RAG) in the legal industry. Acroplans Blog. 2024. URL: https://www.acroplans.com/retrieval-augmented-generation-rag-in-the-legal-industry/ (дата звернення: 22.06.2025)
Published
2025-12-30
How to Cite
Shamov, O. A. (2025). THE ROLE OF GENERATIVE AI IN THE AUTOMATION OF LEGAL DOCUMENT PREPARATION: AN ANALYSIS OF ACCURACY, EFFICIENCY, AND ETHICAL ASPECTS. Systems and Technologies, 70(2), 204-209. https://doi.org/10.32782/2521-6643-2025-2-70.22
Section
COMPUTER SCIENCES