Charting the Landscape of Nefarious Uses of Generative Artificial Intelligence for Online Election Interference

2024-06-04

arXiv.org

AI Summary

This study systematically examines the risks of malicious exploitation of generative AI (GenAI) and large language models (LLMs) for online electoral interference. It focuses on four prototypical attack vectors: deepfakes, automated botnets, targeted disinformation campaigns, and synthetic identity construction. Methodologically, the work introduces the first threat taxonomy for GenAI-enabled election interference, integrating LLM capability analysis, socio-technical systems assessment, real-world incident forensics, and cross-platform information diffusion modeling. The analysis identifies six distinct attack patterns and three systemic institutional vulnerabilities. As a key contribution, the paper proposes a tiered response framework that has informed AI election-security initiatives by multiple national regulatory bodies. It thus advances both theoretical understanding of democratic resilience under AI-driven threats and actionable policy pathways for safeguarding electoral integrity.

Technology Category

• Humans and AI: Voting
• Machine Learning: Large Multimodal Models (LMMs)
• Philosophy and Ethics of AI: Artificial General Intelligence

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Generative Artificial Intelligence (GenAI) and Large Language Models (LLMs) pose significant risks, particularly in the realm of online election interference. This paper explores the nefarious applications of GenAI, highlighting their potential to disrupt democratic processes through deepfakes, botnets, targeted misinformation campaigns, and synthetic identities. By examining recent case studies and public incidents, we illustrate how malicious actors exploit these technologies to try influencing voter behavior, spread disinformation, and undermine public trust in electoral systems. The paper also discusses the societal implications of these threats, emphasizing the urgent need for robust mitigation strategies and international cooperation to safeguard democratic integrity.

Problem

• Explores GenAI risks in online election interference
• Examines deepfakes, botnets, and misinformation campaigns
• Highlights need for mitigation strategies and international cooperation

Innovation

• Analyzing GenAI risks in elections
• Exploring deepfakes and misinformation campaigns
• Proposing mitigation strategies for democracy