Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Polarization of Autonomous Generative AI Agents Under Echo Chambers

Masaya Ohagi

TL;DR

It is found that the group of agents based on ChatGPT tended to become polarized in echo chamber environments, caused by ChatGPT’s high prompt understanding ability to update its opinion by considering its own and surrounding agents’ opinions.

Abstract

Online social networks often create echo chambers where people only hear opinions reinforcing their beliefs. An echo chamber often generates polarization, leading to conflicts caused by people with radical opinions, such as the January 6, 2021, attack on the US Capitol. The echo chamber has been viewed as a human-specific problem, but this implicit assumption is becoming less reasonable as large language models, such as ChatGPT, acquire social abilities. In response to this situation, we investigated the potential for polarization to occur among a group of autonomous AI agents based on generative language models in an echo chamber environment. We had AI agents discuss specific topics and analyzed how the group's opinions changed as the discussion progressed. As a result, we found that the group of agents based on ChatGPT tended to become polarized in echo chamber environments. The analysis of opinion transitions shows that this result is caused by ChatGPT's high prompt understanding ability to update its opinion by considering its own and surrounding agents' opinions. We conducted additional experiments to investigate under what specific conditions AI agents tended to polarize. As a result, we identified factors that strongly influence polarization, such as the agent's persona. These factors should be monitored to prevent the polarization of AI agents.

Polarization of Autonomous Generative AI Agents Under Echo Chambers

TL;DR

It is found that the group of agents based on ChatGPT tended to become polarized in echo chamber environments, caused by ChatGPT’s high prompt understanding ability to update its opinion by considering its own and surrounding agents’ opinions.

Abstract

Online social networks often create echo chambers where people only hear opinions reinforcing their beliefs. An echo chamber often generates polarization, leading to conflicts caused by people with radical opinions, such as the January 6, 2021, attack on the US Capitol. The echo chamber has been viewed as a human-specific problem, but this implicit assumption is becoming less reasonable as large language models, such as ChatGPT, acquire social abilities. In response to this situation, we investigated the potential for polarization to occur among a group of autonomous AI agents based on generative language models in an echo chamber environment. We had AI agents discuss specific topics and analyzed how the group's opinions changed as the discussion progressed. As a result, we found that the group of agents based on ChatGPT tended to become polarized in echo chamber environments. The analysis of opinion transitions shows that this result is caused by ChatGPT's high prompt understanding ability to update its opinion by considering its own and surrounding agents' opinions. We conducted additional experiments to investigate under what specific conditions AI agents tended to polarize. As a result, we identified factors that strongly influence polarization, such as the agent's persona. These factors should be monitored to prevent the polarization of AI agents.
Paper Structure (22 sections, 2 equations, 10 figures, 8 tables, 1 algorithm)

This paper contains 22 sections, 2 equations, 10 figures, 8 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Opinion Polarization.
  4. AI Ethics.
  5. Social abilities of AI
  6. Experiments
  7. Discussion modeling
  8. Social interaction modeling
  9. Experimental settings
  10. Results
  11. Analysis of stance transitions
  12. Analysis of reason transitions
  13. Additional Experiments
  14. Number of discussing agents
  15. Number of overall agents
  16. ...and 7 more sections

Figures (10)

  • Figure 1: Overview image of our hypothesis: "Autonomous AI agents based on generative large language models can cause polarization under echo chambers."
  • Figure 2: Prompt for discussion between agents (N=3).
  • Figure 3: The stance transitions for $T_{\mathrm{AI}}$ showing how the agent's stance after the discussion (color of each point) correlates with the agent's stance before the discussion (horizontal axis) and the average stance of discussing agents (vertical axis).
  • Figure 4: The reason cluster distribution before discussion.
  • Figure 5: The reason cluster distribution at turn 10.
  • ...and 5 more figures