AI assistants spread misinformation in news delivery, study finds
A study coordinated by the European Broadcasting Union and led by the BBC found that AI assistants including ChatGPT, Gemini, Copilot and Perplexity systematically deliver inaccurate news content, with 45% of responses containing serious errors and 20% containing major factual inaccuracies such as fabricated details. Google's Gemini performed worst, with problems in 76% of its responses, primarily due to inadequate source attribution.
What happened
An international study coordinated by the European Broadcasting Union (EBU) and led by the BBC analyzed over 3,000 AI responses from ChatGPT, Copilot, Gemini and Perplexity across 22 public service broadcasters in 18 countries and 14 languages, with professional journalists evaluating the accuracy of news content delivered by the AI assistants.
The study found that 45% of all AI responses contained at least one serious error, 31% had inadequate or misleading source citations, and 20% contained major factual errors such as fabricated details or outdated information. Google's Gemini had problems in 76% of its responses, mainly due to lack of source attribution.
Reported impact
The EBU stated that these shortcomings are systematic, transnational and multilingual, and jeopardize public trust. The organization warned that when people do not know what to trust, they end up trusting nothing, potentially discouraging democratic participation.
Index analysis
A February 2025 BBC study first highlighted AI's problems in handling news. The October 2025 study expanded internationally and confirmed the issue is systemic and not tied to language, market or AI assistant. Comparison between the BBC's earlier results and the expanded study showed some improvements but still high levels of errors.
Governance controls
The EBU and BBC launched a 'News Integrity in AI Assistants Toolkit' to help AI developers and users improve the quality of responses and increase media literacy. The organizations called on the EU and national authorities to apply existing rules on information integrity, digital services and media pluralism, and to introduce ongoing independent review of AI assistants.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.