Original title: Ethics Training Agents: Facilitating Group-Based Ethics Education with Role-Playing and Discussion for Ethical Reflection and Exploration
Youngseok Seo, Sueun Jang, Hyesoo Park, Renz Samuel Gutierrez +2 · September 2026
Teaching ethics to science, technology, engineering, and math students in groups is difficult. It usually takes heavy resources, expert leaders, and manual effort without good online platforms. To fix this, researchers built Ethics Training Agents. It is an online system where people and artificial intelligence models have structured discussions together. The system works by adding multiple artificial intelligence participants that each act out a specific ethical viewpoint, plus an artificial intelligence moderator to guide the group. When tested, the setup supported student engagement, group coordination, and the ability to take on different perspectives. It also had a positive influence on the students' ethical sensitivityThe ability to recognize ethical issues and implications in a situation., showing that artificial intelligence personas can help students explore ethical dilemmas collaboratively.
Imagine a group of friends trying to decide whether to report a classmate for cheating. Each person has a different opinion, but they struggle to see beyond their own view. A neutral teacher steps in, assigning each friend a role: one argues for honesty above all, another for compassion, another for fairness. The teacher ensures everyone listens and responds respectfully. Here, the system does the same: each AI agent represents a distinct ethical stance, and a moderator agent guides the group, helping students explore multiple perspectives they might otherwise miss.
STEM students often struggle to explore ethical dilemmas from multiple perspectives in group settings. Current training is manual and resource-heavy. This system uses AI agents with distinct moral stances and a moderator to guide discussions, making ethics education more accessible, interactive, and scalable without requiring a human expert in every group.
The abstract doesn't say. It covers a user study with 45 undergraduate STEM students and reports positive outcomes on engagement and ethical sensitivity.
Ask anything about this paper. Answers come from its abstract, and say so when the abstract doesn't cover your question.
Ethics Training Agents: Facilitating Group-Based Ethics Education with Role-Playing and Discussion for Ethical Reflection and Exploration — Youngseok Seo, Sueun Jang, Hyesoo Park et al.
Abstract licence: CC0
Explanation generated by Synthesis from the abstract — it is not the original paper. This is a preprint, so it hasn't necessarily been peer reviewed.