AI search shows big differences in source exposure between Baidu and Google
Auditing Source Exposure in Baidu and Google AI Search
Information Retrieval
Summary
AI-generated summaries are common in search engines but behave differently in Chinese compared to English. The authors studied how Baidu and Google show source information in AI answers for similar questions in both languages. They found that the sources visible to users vary a lot between platforms and languages, even when answers seem similar. This means that understanding AI search requires looking both at the answers and who gets credit as the source of information. The study highlights that evaluating AI search results should consider which websites are highlighted, not just the text returned.
What this means in practice
- •For search engine developers: Improve AI answer interfaces by balancing source visibility across languages and platforms based on measured differences in source exposure.
- •For content auditing teams: Audit and adjust AI-generated search results to ensure diverse and representative source exposure in multiple languages and environments.
Authors
Yibo Li, Enci Guan, Yuedan Cai, Geng Liu, Francesco Pierri
Abstract
AI-generated overviews are becoming an increasingly prominent layer of search interfaces, yet their behavior in Chinese-language search remains underexplored. We conduct a cross-lingual audit of AI overview behavior on Baidu and Google using English queries sampled from MS MARCO and their translated Chinese counterparts. Our analysis examines when overviews are triggered across platform-language settings, which host domains receive visible exposure in Chinese-language overviews, how concentrated that exposure is, and how source overlap varies across settings. We also compare the embedding-based semantic similarity of generated answers for matched query intents. The results reveal substantial differences across platform-language settings in overview availability and visible source exposure. At the aggregate level, the settings exhibit low overlap in visible host-domain inventories, while matched-query answers yield median cosine similarities ranging from 0.701 to 0.813. These findings indicate that answer-level semantic similarity and aggregate source exposure capture distinct dimensions of AI-mediated search. Evaluations of AI search should therefore consider not only the content of generated answers but also how source visibility is distributed across platforms, languages, and information environments.