File photo: Leading AI assistants misrepresent news content in almost half of their responses, according to a new study by the European Broadcasting Union (EBU) and the BBC. |Photo provided by: Reuters
Leading AI assistants misrepresent news content in almost half of their answers, according to new research published by the European Broadcasting Union (EBU) and the BBC on Wednesday.
The international study examined 3,000 responses to news questions from leading artificial intelligence assistants (software applications that use AI to understand natural language commands and complete tasks for users).
We evaluated AI assistants, including ChatGPT, Copilot, Gemini, and Perplexity, in 14 languages for their accuracy, source of information, and ability to distinguish between opinion and fact.
Overall, the study showed that 45% of the AI responses examined contained at least one critical issue, and 81% had some issue.
Reuters contacted both companies for comment on the findings.
Gemini, Google’s AI assistant, previously said on its website that it welcomes feedback to continually improve its platform and make it more useful for users.
OpenAI and Microsoft have previously said that hallucinations (when AI models generate false or misleading information, often due to factors such as lack of data) are a problem they aim to solve.
Perplexity says on its website that one of its “Deep Research” modes has a factual accuracy rate of 93.9%.
procurement error
The study found that one-third of the AI assistant’s responses showed significant sourcing errors, such as missing attribution, misleading, or inaccurate information.
About 72% of responses from Gemini, Google’s AI assistant, had serious sourcing issues, compared to less than 25% for all other assistants.
They found accuracy issues with outdated information in 20% of responses from all AI assistants surveyed.
Examples cited in the study include Gemini incorrectly stating changes to the law regarding disposable e-cigarettes and ChatGPT reporting on Pope Francis as the current pope months after his death.
Twenty-two public media organizations from 18 countries participated in the survey, including France, Germany, Spain, Ukraine, the United Kingdom, and the United States.
AI assistants are increasingly replacing traditional news search engines, which could erode public trust, the EBU said.
“When people don’t know what to believe, they end up trusting nothing, which can hinder their participation in democracy,” EBU media director Jean-Philippe de Tender said in a statement.
According to the Reuters Institute’s Digital News Report 2025, around 7% of all online news users and 15% of people under 25 use an AI assistant to get their news.
A new report called for AI companies to take responsibility and improve how their AI assistants respond to news-related questions.
issued – October 22, 2025 10:27 AM IST
