Writing
Posts from 2023 to 2026 first appeared on AI Law Librarians, a group blog I co-founded. They are republished here with their original dates. Subscribe via RSS.
2026
- Six Questions and Three Tests Before You Adopt an AI ToolSix questions and three low-cost tests for deciding whether to adopt an AI tool, use by use, with a worksheet, vendor data-diligence questions, and a sample audition set.
- I Tested Claude for Word on Some Classic Litigator TasksFour tests of the Claude for Word add-in on cite-checking, consistency review, and Bluebooking, with the limits worth knowing before you rely on it.
- Can’t Stop, Won’t Stop: One Semester, Eight Vibe-Coded Teaching ToolsEight custom interactive tools built for one semester of 21st Century Lawyering, from a token explorer to a hallucination game, and what building them taught me.
- What the Science Says About Hallucinations in Legal ResearchPart 1 of a three-part series: six patterns from the research on when and why legal AI hallucinates, and what they mean for practice.
2025
- Thanksgiving Vibe-Coding and the Case for “Single-Serving” Legal SoftwareA Thanksgiving timer app I built instead of asking for a schedule, and why disposable, single-serving software may be the real opportunity in legal practice.
- Legal Research Trapping You in an “AI Tunnel”? Use a Toe-hold to Get OutStudents who start in an AI Ask feature tend to stay there. Three concrete techniques for teaching them to use one good source as a toe-hold out of the tunnel.
- Effortless Boolean: A Free Tool to Supercharge Your Legal ResearchIntroducing the Legal Boolean Search Builder, a free tool that turns my eight-step method for terms-and-connectors queries into a guided app.
- Vibe-Coding Instruction: I Made a Boolean Minigame In 30 MinutesUsing AI coding tools to build a small Boolean-search minigame for legal research class, and what that suggests for interactive instruction.
2024
- RAG Systems Can Still HallucinateA Lexis+ AI answer with two invented cases shows that retrieval-augmented generation reduces hallucinations but does not end them, and when to be most careful.
2023
- Shifting Sands: Ethical Guidance for AI in Legal PracticeA running summary of the first year of ethics guidance on generative AI: bar opinions, court rules, standing orders, and the Fifth Circuit's proposed rule.
- Keeping Up With Generative AI in the LawThe people, newsletters, podcasts, blogs, and LibGuides I rely on to follow generative AI in law, with picks from my co-bloggers.
- The Truth About Hallucinations in Legal Research AI: How to Avoid Them and Trust Your SourcesWhat hallucinations are, why language models produce them, how vendors are trying to contain them, and which legal tasks carry less risk.