Field Notes: Confession of an AI Hallucinator

SHOW NOTESConfessions of an AI Hallucinator: Why Verification Isn’t EnoughIn this Field Note episode, Ron shares a candid story from his early use of ChatGPT — including the moment he nearly relied on hallucinated legal citations in a client memo.This is not just a confession. It’s a practical warning for lawyers tempted to use consumer AI tools for legal research, drafting, and filings without understanding the risks.Ron explains why “just verify it” is not enough, why citation attestations may create a false sense of safety, and why lawyers need something more useful than blanket fear or hype.The answer: Yes, if.Using Ron’s green light / yellow light / red light framework, this episode explores where AI can genuinely help lawyers right now — and where it can absolutely get them into trouble.In the Practice Signal segment, Ron breaks down a lawyer’s question about getting back into FCRA work, and shows how AI could help rebuild a niche practice area from Flintstones → Simpsons → Jetsons. What We Cover•Ron’s near miss with hallucinated case law•Why lawyers get fooled by AI legal output•Why hallucinations are dangerous because they don’t happen every time•Why verification is only the floor, not the ceiling•Why AI citation attestations may not solve the problem•A better “bright line rule” for AI legal drafting•Consumer AI tools vs. legal research platforms•Practice Signal: rebuilding an FCRA practice with AI•FSJ level-up tips for Flintstones, Simpsons, and Jetsons lawyers Key TakeawayDo your legal research first in a trusted legal database. Then use AI to help you think, organize, draft, and improve.AI can absolutely elevate legal work product — but only when it is constrained by verified authority and governed by sound workflow. Resources / Mentions•Westlaw•Lexis+ AI•Bloomberg Law•Fastcase / vLex / Vincent AI•Harvey•Legora•ChatGPT•Claude•Gemini Chapter Markers00:00 – Intro / Field Note setup00:34 – Ron’s confession: the fake case memo03:08 – Why lawyers get fooled by AI legal output04:15 – The real problem: hallucinations don’t happen every time05:21 – Why “verification” is not enough07:08 – AI citation attestations and why they may fail09:16 – Sanctions, contaminated opinions, and court risk10:53 – “Why the hell would I use AI then?”11:18 – The Yes, If framework11:36 – Green light uses for lawyers13:36 – Yellow light uses for lawyers14:03 – Red light uses for lawyers15:25 – Consumer AI vs. legal research tools16:26 – Bright line rule: use the established tool first19:20 – Practice Signal: getting back into FCRA work22:36 – Flintstones / Simpsons / Jetsons level-up tips24:56 – Closing thoughts and call to share