Prompt Engineering and Priming in Law

This guide discusses proper techniques for the use of generative artificial intelligence (GAI) in the context of legal practice and legal method. It highlights the importance of mastering interaction techniques, such as prompt engineering and priming, to unlock the full benefits of GAI while navigating its limitations. The guide discusses several priming techniques that focus on specificity, audience, format, tone, role-based priming, goal-oriented priming, and constraint-based priming. It also delves into advanced prompt engineering methods such as one-shot, few-shot, multi-shot prompting, chain-of-thought prompting, tree-of-thought prompting, maieutic (Socratic) prompting, complexity-based prompting, generated knowledge prompting, least-to-most prompting, self-refine prompting, and directional-stimulus prompting. Additionally, the paper examines the role-playing benefits and constraints in GAI interactions, particularly in legal contexts. The guide provides practical prompt sequences for legal tasks, such as getting grounded in new subjects, understanding complex topics, and drafting legal memoranda, while addressing the challenges of citation accuracy and AI confabulations. It concludes with exercises for users to practice and refine their priming and prompting skills, aiming to enhance their legal practice through effective GAI utilization.