AI voting booth with biased robotic arms

Can AI Truly Represent Us? Unveiling the Biases in AI Voting Systems

Elliot Brynn monogram
Elliot Brynn in Politics & Policy September 2026 • 4 min read.

"Explore how AI language models (LLMs) perform in voting scenarios, revealing biases and limitations in mirroring human choices."


Artificial Intelligence (AI) has made remarkable strides in recent years, particularly in the realm of language processing. Large Language Models (LLMs) like GPT-4 and LLaMA-2 are now capable of understanding and generating human-like text, leading to their integration into various services. However, along with the excitement surrounding these advancements, it's crucial to acknowledge the limitations and potential unforeseen consequences of their widespread use.

One area where LLMs are generating considerable interest is in digital democracy, specifically in assisted real-time voting. The idea of using AI 'digital twins' to represent individual voter preferences has been proposed, raising both enthusiasm and ethical concerns. Proponents believe that AI agents could enable more nuanced and granular voting, but concerns about automation, democratic integrity, and agent bias necessitate careful consideration.

To address these critical questions, a new study investigates the voting behaviors of LLMs, their inherent biases, and how well they align with human voting patterns. By comparing the voting patterns of human participants with those of LLM agents, the study sheds light on the limitations and potentials of integrating LLMs into collective decision-making processes.

AI Search Multiple angles on this topic

The Rise of AI in Political Participation

The intersection of artificial intelligence and democratic participation is a rapidly evolving area of concern and opportunity. As generative AI tools become more capable and widely adopted, their potential role in political decision-making—from shaping public opinion to directly representing voters—has grown significantly. Early research suggests this trend is accelerating, with implications that are not yet fully understood. The precise scale of AI's influence on electoral behavior remains difficult to quantify, but the trajectory is clear and the stakes are high.

LLMs as Voter Proxies: Promise and Pitfalls

Recent advances in generative AI and large language models (LLMs) present a significant opportunity to expand democratic participation. These models could help overcome cognitive bandwidth limitations by providing decision support or even acting as direct representatives for abstaining voters at scale. However, researchers caution that the quality of this AI representation and the underlying biases it introduces remain open questions. The core tension lies in balancing the potential for fairer outcomes with the risk of encoding AI-driven inconsistencies into collective choice.

From Voting Reform to AI Assistants

Efforts to improve democratic participation through better voting systems are not new. Scholars have long debated the merits of alternative voting methods designed to yield fairer outcomes. What is new is the prospect that AI could automate or augment these processes, potentially transforming the landscape of collective decision-making. This represents a paradigm shift from purely human-driven democratic innovation to technology-assisted participation.

Do AI Voting Systems Mirror Human Choices?

AI voting booth with biased robotic arms

The study, titled 'LLM Voting: Human Choices and AI Collective Decision-Making,' explores the voting behaviors of Large Language Models (LLMs), specifically GPT-4 and LLaMA-2. Researchers used a dataset from a human voting experiment to establish a baseline for human preferences. They then conducted a corresponding experiment with LLM agents, analyzing their choices and biases.

The researchers observed that LLM voting outcomes were influenced by the choice of voting methods and the presentation order of candidates. This highlights a key challenge: LLMs are susceptible to biases that can skew their decisions, much like how the arrangement of candidates on a ballot can influence human voters.

  • Voting Method Matters: The way votes are cast (e.g., approval voting, ranked voting) significantly impacts LLM choices.
  • Presentation Order: The order in which options are presented affects LLM voting outcomes.
  • Persona Influence: Adopting different personas can reduce biases and improve alignment with human choices.
  • Chain-of-Thought Potential: While not improving accuracy, the 'Chain-of-Thought' approach shows promise for AI explainability.
AI Search Multiple angles on this topic

Mapping AI's Role in Democratic Health

The relationship between AI and democracy is the subject of vigorous scholarly debate. Some researchers argue that AI expansion can enhance democratic health by improving public services, fostering citizen trust, and boosting economic opportunities. Conversely, others contend that AI poses substantial threats to democratic values, including privacy, equality, and informed participation. The landscape of this debate spans multiple dimensions, from election integrity to the quality of public discourse.

AI's Democracy Problem and Racial Harms

A critical examination of AI in politics rests on the premise that democracy itself functions as an information system—one that AI can both support and distort. Research highlights that AI models, by defaulting to averages or dominant patterns, often fail to recognize diverse perspectives and needs, including those of racial and ethnic minorities. Generative AI also acts as a powerful accelerator for citizen-initiated direct democracy mechanisms, automating law-drafting and enabling hyper-personalized persuasion. While this efficiency lowers historical barriers, it simultaneously threatens the deliberative quality that gives democratic outcomes their legitimacy.

AI vs. AI: Boundary Conditions for Democratic Use

Researchers have begun delineating the ethical and practical boundaries of using AI systems to counter other AI systems in democratic contexts. This 'AI versus AI' approach raises important questions about the conditions under which such competition can genuinely support deliberative decision-making. Studies also examine whether fair collective choice outcomes can be resilient to the biases inherent in different LLM-based voting assistants. The emerging consensus is that the factors influencing outcomes must be carefully considered before deploying such systems at scale.

One significant finding was the trade-off between preference diversity and alignment accuracy in LLMs. Different temperature settings, which control the randomness of the LLM's responses, influenced this balance. The study indicates that LLMs may lead to less diverse collective outcomes and biased assumptions when used in voting scenarios. This underscores the need for cautious integration of LLMs into democratic processes.

The Future of AI in Democratic Decision-Making

The study's findings serve as a reminder that while LLMs offer exciting possibilities, their integration into democratic processes requires careful consideration. As AI continues to evolve, addressing these biases and limitations will be crucial to ensuring fair and representative outcomes in collective decision-making.

AI Search Multiple angles on this topic

Political Bias in AI Chatbots as a Global Concern

As voters increasingly turn to large language models for political tasks, including voting guidance, the political neutrality of AI chatbots has become a major policy concern. Researchers note that American AI chatbots are used globally, yet little is known about their political behavior and potential biases in non-U.S. contexts. Expert analysis also identifies AI as a marketing term that encompasses data-intensive analysis, predictions, and surveillance serving industry and power. Meanwhile, election risks amplified by AI—including disinformation, vote suppression, and security hazards—long predate the generative-AI boom, suggesting that legislative reforms beyond AI-specific measures are necessary.

Harnessing AI to Strengthen Democracy

Governments, companies, and civil society must collaborate to address the limitations of existing regulation in the face of AI-driven deception. While many experts fear AI will be deployed to weaken democracy, examples exist of it being used to make democratic systems fairer and more inclusive. Scholars note that AI is just the latest in a long line of technologies that have influenced politics throughout history. The challenge going forward is ensuring that democratic institutions adapt quickly enough to harness AI's benefits while mitigating its risks.

Institutional and Structural Dimensions

The implications of AI for democratic systems extend beyond individual voting behavior to encompass broader institutional and structural challenges. These include questions about accountability, transparency, and the distribution of power in societies increasingly shaped by algorithmic decision-making. Any assessment of AI's impact on democracy must account for these systemic factors, which are difficult to capture in narrowly focused studies.

What Democracy Means for Real People

Ultimately, the debate about AI and democracy is not merely academic—it concerns the lived experiences of citizens navigating an increasingly complex information environment. The human element in democratic participation, including trust, engagement, and the capacity for informed deliberation, remains central even as technology reshapes the landscape. How societies manage this transition will determine whether AI becomes a tool for genuine democratic empowerment or a source of deeper disengagement and inequality.

About this Article -

Written with AI assistance from published research, and reviewed by the Mystum team. See our About page for more information.

This article is based on research published under:

DOI-LINK: 10.1609/aies.v7i1.31758,

Title: Llm Voting: Human Choices And Ai Collective Decision Making

Subject: cs.cl cs.ai cs.cy cs.lg econ.gn q-fin.ec

Authors: Joshua C. Yang, Damian Dailisan, Marcin Korecki, Carina I. Hausladen, Dirk Helbing

Published: 31-01-2024

Everything You Need To Know

1

What are the primary Large Language Models (LLMs) examined in the study regarding their voting behaviors?

The study focuses on the voting behaviors of two prominent Large Language Models (LLMs): GPT-4 and LLaMA-2. These models were chosen to represent the capabilities of current AI in understanding and generating human-like text. The research explores how these specific LLMs perform in mimicking human voting patterns and the biases they exhibit in decision-making processes.

2

How do different voting methods impact the decision-making of Large Language Models (LLMs), and what are the implications?

The research highlights that the choice of voting methods significantly influences the outcomes generated by LLMs. Different methods, such as approval voting and ranked voting, can lead to varied results, mirroring how these methods affect human voters. This suggests that the specific voting system used can introduce bias and affect the alignment of LLM choices with human preferences. It underscores the necessity of carefully selecting voting methods to ensure fairness and prevent skewed outcomes when integrating LLMs into democratic processes. The study also finds that the presentation order of the candidates also affects the decision of the LLMs.

3

Can adopting different personas improve the accuracy or fairness of Large Language Models (LLMs) in voting scenarios?

Yes, the study suggests that adopting different personas can reduce biases and improve the alignment of LLMs with human choices. By giving the LLMs different 'personalities' or perspectives, researchers found that the models' voting patterns could be altered to better reflect human preferences. This indicates a potential method for mitigating bias, although it also introduces complexities in determining which personas are most representative or desirable in a voting context. The researchers also explored 'Chain-of-Thought' to understand the decision of LLMs.

4

What is the trade-off between preference diversity and alignment accuracy when using Large Language Models (LLMs) in voting scenarios?

The study reveals a trade-off between preference diversity and alignment accuracy in LLMs. Different temperature settings, which control the randomness of the LLM's responses, influence this balance. Higher temperatures can lead to more diverse but potentially less accurate outcomes, whereas lower temperatures might result in more aligned but less diverse results. This means that when LLMs are used in voting, there's a risk that the collective outcomes could be less diverse than human preferences, especially if the LLM is tuned for high accuracy. This highlights the need for careful calibration and consideration of how LLMs are used in collective decision-making to ensure fair and representative outcomes.

5

What are the key challenges and ethical concerns of integrating Large Language Models (LLMs) into digital democracy and voting systems?

The integration of Large Language Models (LLMs) into digital democracy raises several key challenges and ethical concerns. One major concern is the potential for bias in LLMs, which can skew voting outcomes and misrepresent voter preferences. The study highlights that LLMs are susceptible to influences like voting methods, presentation order, and even the adoption of different personas. Other concerns include the need to ensure democratic integrity, prevent automation bias, and carefully consider how LLMs' inherent limitations might affect collective decision-making. Also, the opacity of LLM decision-making, and the potential for manipulation or misuse of these systems, further complicate the ethical considerations, underscoring the need for cautious and thoughtful integration of LLMs into democratic processes.

Newsletter Subscribe

Subscribe to get the latest articles and insights directly in your inbox.