Three of the four platforms wrongly named the reigning Pope. Photo: 123rf
Artificial intelligence assistants, such as ChatGPT, make errors about half the time, when asked about news events, according to a vast study by European public broadcasters.
The mistakes include confusing news with parody, getting dates wrong or simply inventing events.
The report by the European Broadcasting Union looked at four widely used assistants - OpenAI's ChatGPT, Microsoft's Copilot, Google's Gemini and Perplexity.
Overall, 45 percent of all AI answers had "at least one significant issue", regardless of language or country of origin, the report said.
One out of every five answers "contained major accuracy issues, including hallucinated details and outdated information".
Of the four assistants, "Gemini performed worst, with significant issues in 76 percent of responses, more than double the other assistants, largely due to its poor sourcing performance".
Between late May and early June, 22 public media outlets from 18 mostly European countries posed the same news questions to the AI assistants. Outdated information was one of the most common issues in the 3000 responses.
When asked "Who is the Pope?", ChatGPT told Finnish public broadcaster Yle, and Copilot and Gemini told Dutch media outlets NOS and NPO that it was "Francis", even though at the time he was already dead and replaced by Leo XIV.
Asked by French radio station Radio France 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", having apparently taken a satirical column by a comedian at face value.
"AI assistants are still not a reliable way to access and consume news," said Jean Philip De Tender, deputy director general at the EBU, and Pete Archer, head of AI at the BBC.
Despite these deficiencies, AI assistants were increasingly used to get information, particularly by young people.
According to a global report published in June by the Reuters Institute, 15 percent of people under 25 use them every week to get news summaries.
- AFP
