An AI architecture combining a language model with a retrieval system that fetches relevant documents at query time, grounding responses in authoritative source material to reduce hallucination.
Last reviewed: 2026/05/18
An AI architecture combining a language model with a retrieval system that fetches relevant documents at query time, grounding responses in authoritative source material to reduce hallucination. Define retrieval-augmented generation for legal teams, then connect grounding concepts to research and document workflows.
An AI-driven multi-step legal process — such as intake to routing to drafting — that executes autonomously across defined stages without per-step human prompting.
Tech / ModelA quantitative measure of how often an AI system produces correct outputs on a defined test set — critical for evaluating legal AI tools where errors carry professional responsibility risk.
Tech / ModelA standardized documentation artifact describing an AI model's intended use, performance characteristics, limitations, and training data — essential for legal AI vendor due diligence.
Tech / ModelA standardized evaluation measuring an AI system's accuracy, reliability, or performance on defined legal tasks — used to compare tools and validate fitness for professional use.
The most expensive legal AI in the market — Am Law 100 firms only.
Thomson Reuters' GPT-backed legal research and drafting with Westlaw integration (relaunched as CoCounsel Legal, 2025).
Purpose-built US legal AI covering research, drafting, and compliance.
Move from this definition to role-based legal AI shortlists and the selection criteria that matter for each type of legal team.
Am Law 200 and global firm workflows: accuracy at scale, security compliance, and matter-level auditability.
60 legal AI tools vetted for the solo lawyer: tight budget, no IT team, billable-hours pressure.
Last reviewed: 2026/05/18. Definitions are written by the LawyerAI Editorial team. Commercial relationships are disclosed and do not determine editorial scores or conclusions. See our Sponsorship & Affiliate Disclosure.
Retrieval-Augmented Generation (RAG) is an AI architecture in which a language model is paired with a search or retrieval component. When a user submits a query, the system first retrieves relevant documents — such as statutes, case law, contract clauses, or internal precedents — and supplies them as context to the language model before it generates a response. This approach grounds the model's output in specific, verifiable source material rather than relying solely on knowledge encoded during training, which can be outdated or incomplete.
Legal work demands citation to authoritative sources, and the core risk of unaided language models — generating plausible but fabricated case citations or statutory text — is substantially reduced when RAG is used with a properly curated legal corpus. For law firms and legal departments, understanding whether a vendor uses RAG (and what documents populate its retrieval index) is a practical competence and accuracy question. RAG also enables AI systems to be updated with new law without retraining the underlying model.