AI-Powered Lawyering: AI Reasoning Models, Retrieval Augmented Generation, and the Future of Legal Practice

Юридическая работа с использованием ИИ: модели рассуждения ИИ, генерация с дополнением извлечёнными данными и будущее юридической практики
Daniel B. Schwarcz, Sam Manning, J.C. Prescott, Patrick Barry, David R. Cleveland, Beverly Rich
2026-04-09

AI reasoning modelslegal AIlegal hallucinationsrandomized controlled trialretrieval-augmented generation
Generative AI is set to transform the legal profession, though its most promising uses and ultimate effects are still unclear. While AI models like GPT-4 improve efficiency, they can also “hallucinate” and may undermine legal judgment, particularly in complex tasks typically handled by skilled lawyers. This article examines two emerging AI innovations that may mitigate these concerns: Retrieval Augmented Generation (RAG), which grounds AI-powered analysis in legal sources, and AI reasoning models, which structure complex reasoning before generating output. We conduct the first randomized controlled trial assessing these technologies, assigning upper-level law students to complete legal tasks using a RAG-powered legal AI tool (Vincent AI, 2024), an AI reasoning model (OpenAI’s o1-preview), or no AI. We find that both AI tools significantly enhance legal work quality, a marked contrast with previous research examining older large language models like GPT-4. Moreover, these newer models appear to maintain the efficiency benefits associated with older AI technologies. Our findings also show that these AI tools significantly boost productivity in five out of six tested legal tasks, with statistically significant gains of anywhere from 50% to 130%. They perform exceptionally well in complex tasks like drafting persuasive letters and analyzing complaints. Notably, o1-preview improves the analytical depth of work product and Vincent AI avoids introducing more hallucinations, suggesting that integrating domain-specific RAG capabilities with reasoning models could yield even larger improvements.
1
AI assistance significantly increases productivity in five of six legal tasks, with gains ranging from 50% to 130%.
2
The newer AI tools preserve earlier generative AI efficiency benefits while avoiding the quality limitations observed in prior GPT-4 research.
3
The randomized controlled trial finds that both RAG-powered legal AI and an AI reasoning model significantly improve the quality of law students’ legal work compared with no AI.
4
The tools perform especially well on complex legal tasks, including drafting persuasive letters and analyzing complaints.
5
o1-preview produces greater analytical depth, while Vincent AI does not introduce more hallucinations; combining retrieval augmentation with reasoning models may yield larger improvements.

AI-powered legal practice using Retrieval Augmented Generation and AI reasoning models

The effects of RAG and AI reasoning models on the quality, productivity, efficiency, analytical depth, and hallucination rate of legal work

Publication Details
Publication Date
2026-04-09
Journal
Publisher
ISSN
Cited by
47
Access Type
Author Information
Authors
Daniel B. Schwarcz
Sam Manning
J.C. Prescott
Patrick Barry
David R. Cleveland
Beverly Rich
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%