How Does Generative Artificial Intelligence Gatekeep News? A Cross-National Comparative Study of the Perplexity Platform
Generative artificial intelligence is emerging as a new gateway for users to access news. Using the Perplexity platform as the research subject, this study draws on full-sample data from its news gateways across 15 countries and regions between January and February 2026 to examine how generative AI news gateways perform news gatekeeping across the three dimensions of sources, issues, and national context. The findings reveal that Perplexity's news sources are highly concentrated among a small number of global general-interest news organizations, reflecting the continued dependence of news practices on institutional authority. The source structure varies by topic type. Geopolitical issues rely heavily on general news sources, while technology issues incorporate more specialized sources, closely mirroring the institutional patterns that have long shaped the relationship between topics and sources in traditional news production. News gateways across countries draw on a shared core of news organizations, while source weights and the relationship between topic and source composition vary across countries. Generative gatekeeping, jointly driven by technology and platforms, is thus becoming deeply embedded in existing institutions of news production and the global landscape of international news flows.
Authors
- Jinpeng Wang (ORCID: https://orcid.org/0000-0002-4041-4633)
Institutions
- Tsinghua University (CN)
Publication Details
- Journal
- Media Technology and Governance
- Published
- 2026-11-18
- DOI
- https://doi.org/10.1177/29777984261506956
- Primary Topic
- News Media Framing and Political Communication
- Type
- article
- Field-Weighted Citation Impact
- 0.00