Breaking the bubble: How can generative AI reshape opinion formation and echo chambers

Abstract

The rapid advancement of Generative Artificial Intelligence (GAI) is reshaping opinion formation by influencing how individuals consume and interpret information. AI-driven tools such as ChatGPT, Copilot, and DeepSeek offer access to diverse viewpoints, yet they also risk reinforcing echo chambers and filter bubbles, limiting exposure to alternative perspectives. This study examines the role of GAI in shaping, reinforcing, or disrupting digital silos, focusing on the interaction between AI generated content and human cognition.

This study adopts a qualitative design using semi-structured interviews with industry experts and GAI users to explore how generative AI tools influence opinion formation. Thematic analysis is applied to uncover key patterns in user engagement and AI’s role in digital discourse. The findings contribute to theoretical debates on opinion formation by integrating human-based cognitive models with AI driven learning mechanisms. Managerially, this study offers insights for AI developers, policymakers, and marketers on designing AI systems that encourage balanced information consumption. By questioning AI’s potential as either a neutral facilitator or a biased curator, this study contributes to broader discussions on AI ethics, misinformation, and digital polarisation in the evolving media landscape.