Answer Bubbles: Information Exposure in AI-Mediated Search
This research identifies potential 'answer bubbles' in AI-mediated search that could impact user trust and information transparency, representing an incremental analysis of existing systems.
The study analyzed 11,000 search queries across four systems to compare generative AI summaries with traditional search, finding that AI systems exhibit significant source-selection biases, such as overrepresenting Wikipedia by up to 60% and underrepresenting social media, and reduce hedging language while preserving confidence.
Generative search systems are increasingly replacing link-based retrieval with AI-generated summaries, yet little is known about how these systems differ in sources, language, and fidelity to cited material. We examine responses to 11,000 real search queries across four systems -- vanilla GPT, Search GPT, Google AI Overviews, and traditional Google Search -- at three levels: source diversity, linguistic characterization of the generated summary, and source-summary fidelity. We find that generative search systems exhibit significant \textit{source-selection} biases in their citations, favoring certain sources over others. Incorporating search also selectively attenuates epistemic markers, reducing hedging by up to 60\% while preserving confidence language in the AI-generated summaries. At the same time, AI summaries further compound the citation biases: Wikipedia and longer sources are disproportionately overrepresented, whereas cited social media content and negatively framed sources are substantially underrepresented. Our findings highlight the potential for \textit{answer bubbles}, in which identical queries yield structurally different information realities across systems, with implications for user trust, source visibility, and the transparency of AI-mediated information access.