LLMs for Lawyers

Topic · 7 guides

LLM Fundamentals for Lawyers

How large language models work, why they hallucinate, what context windows and RAG mean, and how to read benchmarks - the mental model every other cluster assumes.

Start hereHow Large Language Models Actually Work (Explained for Lawyers)How LLMs work, explained for lawyers: tokens, training, context windows and why ChatGPT invents cases. The mental model to build before you prompt.16 min read →
  1. AI Sycophancy: Why the Model Agrees With Everything You Say (and How to Make It Argue Back)LLMs are trained to please. Why ChatGPT and Claude validate weak arguments, how sycophancy corrupts legal analysis, and prompts that force pushback.9 min read
  2. RAG Legal Research Explained: Why 'Grounded in Westlaw' Still Gets the Law WrongRAG legal research explained: retrieval promised hallucination-free answers; Stanford found 17-33% errors. What it fixes, what it cannot, how to use it.10 min read
  3. The AI Glossary for Lawyers: 40 Terms, Each With a Legal ExampleAn AI glossary for lawyers: token, context window, RAG, hallucination, MCP, agent, zero data retention and 33 more terms, each with a legal example.10 min read
  4. The Context Window Explained for Lawyers: How Much Should You Upload Before Context Rot Sets In?Context window explained for lawyers: can you upload a 300-page contract? What context rot does to long documents, and how to feed files to AI safely.9 min read
  5. The Legal AI Benchmarks, Decoded: What Was Tested, Who Refused, and What the Numbers MeanWhat legal AI benchmarks measure: Stanford's hallucination studies, Vals VLAIR, LinksAI, BigLaw Bench and LEXam, who refused testing, and what it means.11 min read
  6. Why Does AI Make Up Fake Cases? Hallucination Explained for LawyersWhy does AI make up fake cases? How ChatGPT, Claude and Lexis+ AI hallucinate citations: the statistics, mechanics and traps to know before you file.11 min read

Other topics