BBC–EBU Study Unveils Major AI Accuracy Issues
A recent study by BBC and the European Broadcasting Union (EBU) exposed deep flaws in the accuracy of AI-generated news responses. The research found that 45% of answers provided by leading AI models were inaccurate. Among the tested systems, Gemini performed the worst, raising concerns about its reliability for delivering factual news.
Gemini Lags Behind in Reliability and Sourcing
Gemini’s performance in the study highlights ongoing issues with sourcing and factual accuracy in AI-powered news answers. The report emphasizes that, while AI tools have become more popular for information retrieval, users should remain cautious and verify content from multiple sources. The findings also call for tech companies to prioritize improvements in AI accuracy, transparency, and citation of reputable sources.
Sources:
https://www.digitalinformationworld.com/2025/10/gemini-struggles-most-in-accuracy-test.html
