LLM Voting: Human Choices and AI Collective Decision-Making


METADATA ONLY
Loading...

Date

2024

Publication Type

Conference Paper

ETH Bibliography

yes

Citations

Altmetric
METADATA ONLY

Data

Rights / License

Abstract

This paper investigates the voting behaviors of Large Language Models (LLMs), specifically GPT-4 and LLaMA-2, their biases, and how they align with human voting patterns. Our methodology involved using a dataset from a human voting experiment to establish a baseline for human preferences and conducting a corresponding experiment with LLM agents. We observed that the choice of voting methods and the presentation order influenced LLM voting outcomes. We found that varying the persona can reduce some of these biases and enhance alignment with human choices. While the Chain-of-Thought approach did not improve prediction accuracy, it has potential for AI explainability in the voting process. We also identified a trade-off between preference diversity and alignment accuracy in LLMs, influenced by different temperature settings. Our findings indicate that LLMs may lead to less diverse collective outcomes and biased assumptions when used in voting scenarios, emphasizing the need for cautious integration of LLMs into democratic processes.

Publication status

published

Book title

Proceedings of the Seventh AAAI/ACM Conference on AI, Ethics, and Society (AIES-24)

Volume

7 (1)

Pages / Article No.

1696 - 1708

Publisher

AAAI

Event

7th AAAI/ACM Conference on AI, Ethics, and Society (AIES 2024)

Edition / version

Methods

Software

Geographic location

Date collected

Date created

Subject

Organisational unit

03784 - Helbing, Dirk / Helbing, Dirk check_circle

Notes

Funding

187249 - Decisionmaking process supported by a participatory platform: Consequences on trust, on legitimacy of the political decision, and user skills. (SNF)
833168 - Co-Evolving City Life (EC)
953783 - Distributed Intelligence and Technology for Traffic and Mobility Management (EC)

Related publications and datasets