Language

English (en)

Date Uploaded

Fall 9-4-2026

Activity source

Original

Summary

Students conduct a controlled, two-run generative AI jury simulation to examine how a single socio-legal variable affects simulated jury deliberation. They compare and manually code the AI-generated outcomes against peer-reviewed human jury research to evaluate the model’s accuracy, limitations, and potential biases.

Extended Summary

In this activity, students use generative AI as a simulated behavioral data generator to investigate factors that influence jury deliberation and decision-making. After reviewing foundational research and selecting a criminal case scenario, students choose one socio-legal variable grounded in peer-reviewed jury research, conduct a baseline AI-generated jury deliberation, and then conduct a second simulation in which only that variable is changed. Students manually code and compare the two deliberation transcripts using a structured behavioral taxonomy, then evaluate their findings against published research involving human jurors. The activity culminates in a synthesis report in which students assess whether the AI-generated patterns are consistent with empirical research, identify stereotypes or other limitations in the model’s behavior, and consider the ethical implications of using AI-based predictive tools in high-stakes legal contexts.

Student Learning Objectives

By completing this activity, students will be able to:

  • Explain how socio-legal variables can influence jury deliberation and decision-making.
  • Apply findings from peer-reviewed jury research to the analysis of simulated jury behavior.
  • Design a controlled comparison by manipulating one research variable while holding other simulation parameters constant.
  • Systematically code and compare behavioral patterns across AI-generated deliberation transcripts.
  • Evaluate the extent to which generative AI outputs are consistent with or diverge from findings in human jury research.
  • Identify potential stereotypes, biases, oversimplifications, and other limitations in AI-generated representations of human behavior.
  • Critically evaluate the ethical implications of using AI and algorithmic prediction in high-stakes legal and forensic contexts.

Assignment Type

Both in-class and out-of-class

Course level

Multiple Course Levels Apply

Used in course?

yes

If yes, enter name of course in box

Forensic Psychology

Type of Student-AI Collaboration Required

Collaboratively generated by human and AI

Type of AI Task(s)

Textual Analysis

Second Type of AI Task

Recommendation/Decision-Making

Third Type of AI Task

Data Analysis

Uploader/Author Affiliation

Faculty

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Document Type

Teaching Material

Download

Share

COinS