Document Type
Article
Publication Title
Washington University Journal of Law and Policy
Abstract
Forthcoming in 79 Wash. U. J. of Law and Policy ____ (2025)
AI “bots” like ChatGPT can negotiate without training, but they are overly cooperative, making concessions that weaken their bargaining position—unlike skilled legal negotiators, who adapt their tactics to individual situations as well as what occurs in the process and move as needed between cooperative and competitive tactics.
The article explores how to train bots to “negotiate like lawyers” by giving them role instructions, knowledge about bargaining, and iterative practice, including step-by-step instructions to create a bargaining bot in ChatGPT.
The author notes that texts on negotiation, and his own teaching methods, do not give students specific advice about what to do in different bargaining situations, and describes how the process of training a bot prompted him to explore his own often unconscious “rules of bargaining.” The author concludes that AI bots offer exciting new tools to deepen our understanding of bargaining and improve negotiation outcomes.
Publication Date
7-25-2025
Recommended Citation
Golann, Dwight, "Can We Train Bots to Negotiate Like Lawyers? And What Do We Learn About Bargaining as We Do?" (2025). Suffolk University Law School Faculty Works. 307.
https://dc.suffolk.edu/suls-faculty/307
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License
Included in
Dispute Resolution and Arbitration Commons, Legal Profession Commons, Other Law Commons, Science and Technology Law Commons