Did My AI Just Make That Up? A Defensive Prompting Toolkit for Hallucination-Proofing Legal Work

As generative AI becomes further embedded in legal research, drafting, and contract review, fabricated citations and false authority present a very real risk for legal practitioners. This program is a focused session in defensive prompting: the discipline of prompting to protect your work product, not just accelerate it. The presenter will share practical techniques used to measurably reduce AI hallucinations in legal work. Drawing from real-world prompting frameworks and legal workflows, the session will cover constraint-based instructions, source-grounding methods, and verification processes that deliver more reliable, accurate outputs. Attendees will leave with a concrete defensive prompting toolkit they can put to use immediately, so AI stays a reliable assistant, not a source of professional risk.