UNT Dallas Law Review On The Cusp / SSRN // August 22, 2025
Michael Scott Is Not a Juror: The Limits of AI in Simulating Human Judgment
Examines why current LLMs do not reliably replicate human juror reasoning and where controlled jury simulation may still be useful.
Summary
This paper pushes back on the idea that large language models can stand in for actual jurors. It separates surface-level plausibility from the social, experiential, and deliberative features of human judgment.
At the same time, it identifies narrower research uses for simulation, provided the limits are explicit and the outputs are treated as inputs for further analysis rather than substitutes for real-world decision makers.
Original publication
The full article lives at its original source. Use the outbound link below to open the complete publication.