Social Sciences · General Social Sciences
Researchers used large language models to create artificial survey respondents for a weekly look at the 2026 US House elections. Adding personal memories and news exposure shifted the estimates toward Democrats, but the resulting 18.7-point Democratic margin was well above the 8.7-point polling average, leaving the method’s predictive accuracy unresolved.
A study tested whether an AI language model could distinguish factual accuracy from the appearance of credibility. The model often gave fabricated texts fairly high reliability scores, although a separate test of whether claims matched external facts performed better.
A review of 121 studies finds that artificial intelligence is changing journalism through automation and human–machine collaboration, but its legal and social rules remain unsettled. Researchers identify disclosure practices, cultural context and copyright as central issues for future study.
