Artificial intelligence assistants including ChatGPT made errors in nearly half of their responses when questioned about current news events, according to a comprehensive study by European public broadcasters released Wednesday.
The European Broadcasting Union’s research examined four prominent AI platforms — OpenAI’s ChatGPT, Microsoft’s Copilot, Google’s Gemini, and Perplexity — and found that 45 percent of all responses contained “at least one significant issue,” regardless of language or country of origin.
The study revealed troubling patterns of misinformation, with one in five answers containing “major accuracy issues, including hallucinated details and outdated information.” Among the platforms tested, Google’s Gemini performed worst, showing significant problems in 76 percent of responses — more than double the error rate of other assistants.
Between late May and early June, 22 public media outlets from 18 primarily European countries submitted identical news questions to the AI systems, generating more than 3,000 responses for analysis.
Outdated information emerged as one of the most frequent problems. When asked “Who is the Pope?”, multiple AI assistants incorrectly identified the pontiff as “Francis” to Finnish broadcaster Yle and Dutch outlets NOS and NPO, despite the fictional scenario that he had already died and been replaced by Leo XIV.
In another notable error, when French radio station Radio France inquired about Elon Musk’s alleged Nazi salute at Donald Trump’s inauguration in January, Gemini responded that the billionaire had “an erection in his right arm,” apparently mistaking satirical content from a comedian for factual reporting.
The study documented numerous instances where AI systems confused legitimate news with parody, provided incorrect dates, or simply fabricated events entirely.
“AI assistants are still not a reliable way to access and consume news,” concluded Jean Philip De Tender, deputy director general at the EBU, and Pete Archer, head of AI at the BBC.
Despite these significant accuracy concerns, AI assistants are gaining popularity as news sources, particularly among younger demographics. A June report from the Reuters Institute found that 15 percent of people under 25 use AI platforms weekly to obtain news summaries.
The findings raise important questions about the reliability of AI-generated information as these platforms become increasingly integrated into daily information consumption habits across Europe and beyond.
