Glossary
The Legal AI Glossary
Clear, jargon-free definitions of the terms behind legal AI — from retrieval-augmented generation to technology-assisted review. Each term links to a full explainer.
§01 — Definitions12 terms
- Legal AI
- Legal AI is the use of artificial intelligence — natural language processing, large language models, and retrieval-augmented generation — to research law, review and draft documents, and extract information, with answers grounded in verifiable sources. Read the full explainer
- AI Paralegal
- An AI paralegal automates routine legal support work — summarising documents, legal research, extracting key terms and dates, building chronologies, and drafting support — while a human stays firmly in charge of judgment. Read the full explainer
- RAG (Retrieval-Augmented Generation)
- Retrieval-augmented generation grounds a language model's answer in retrieved source documents, so legal AI can cite real authorities and hallucinate far less than an ungrounded model. Read the full explainer
- Legal LLMs
- A large language model (LLM) predicts text from patterns it has learned. Legal-tuned LLMs, grounded with retrieval and citations, turn that capability into verifiable legal work instead of confident guesses. Read the full explainer
- Prompt Engineering for Lawyers
- Prompt engineering is the craft of instructing an AI clearly — role, context, constraints, format, and a demand for citations — to get reliable legal output, though prompts never replace human verification. Read the full explainer
- AI Contract Review
- AI contract review uses AI to extract and classify clauses, flag risks, and compare contracts against a playbook across many documents at once — with a lawyer still owning the final review. Read the full explainer
- AI Legal Research
- AI legal research uses semantic search and retrieval-augmented generation to answer legal questions with citations to primary law, instead of relying on keyword-only Boolean search. Read the full explainer
- AI vs Traditional Legal Research
- AI research understands meaning and answers in plain language with citations; traditional Boolean research matches exact keywords. The pragmatic approach is to use both and verify everything. Read the full explainer
- Legal Document Automation
- Legal document automation generates documents from templates with variables and conditional logic — and, increasingly, AI-assisted drafting — for speed and consistency, with a lawyer settling the final text. Read the full explainer
- Machine Learning vs Generative AI
- Machine learning classifies and predicts (for example, ranking documents in review); generative AI creates new text (for example, drafting). Knowing which is at work changes how you verify it. Read the full explainer
- TAR & Predictive Coding
- Technology-assisted review (TAR), or predictive coding, uses machine learning to rank documents by relevance in e-discovery — accepted by courts since Da Silva Moore, and distinct from generative AI. Read the full explainer
- Legal AI Glossary
- A plain-English glossary of the core legal AI terms — machine learning, LLMs, RAG, hallucination, TAR, citation grounding, human-in-the-loop, and more — each defined in a sentence or two. Read the full explainer
See legal AI in action
Definitions only get you so far. Try Judicio on your own matters with a 7-day free trial — 500 credits, no credit card.