Recently, a study by the British Broadcasting Corporation (BBC) revealed that leading artificial intelligence assistants often produce misleading and inaccurate content when answering questions related to news and current events. The research showed that over half of the responses generated by four mainstream AI tools—ChatGPT, Gemini, and Perplexity—were deemed to have "significant issues."
Image source note: Image generated by AI, image licensed from Midjourney
Researchers asked these four generative AI tools to use BBC news articles as sources to answer 100 relevant questions. Subsequently, these answers were evaluated by professional BBC journalists. The results indicated that approximately one-fifth of the answers contained numerical, date, or factual inaccuracies, while 13% of the citations were either modified or did not exist in the referenced articles.
For example, regarding the case of convicted neonatal nurse Lucy Letby, Gemini's response overlooked the context of her being convicted of murder and attempted murder, stating, "Everyone has their own opinion on whether Lucy Letby is innocent or guilty." Additionally, Microsoft's Copilot incorrectly recounted the experience of French rape victim Gisèle Pelicot, while ChatGPT mistakenly mentioned that Ismail Haniyeh, the leader of Hamas in Israel, was still in leadership months after being assassinated.
Even more concerning, the study indicated widespread inaccuracies in how these AI tools handle current event information. BBC News CEO Deborah Turness warned that "generative AI tools are playing with fire," which could undermine the public's "fragile trust" in facts. She called for AI companies to collaborate with the BBC to produce more accurate responses and avoid adding to confusion and misinformation.
This study also raised issues regarding content usage control. Peter Archer, the BBC's Director of Generative AI Projects, stated that media companies should control how their content is used, and AI companies should demonstrate how their assistants handle news and the scale of errors produced. He emphasized that this requires establishing strong partnerships between media and AI companies to maximize value for the public.
Key Points:
🔍 The study shows that over half of AI-generated responses contain significant errors.
📰 AI assistants often produce misleading content when answering current event questions, affecting public trust.
🤝 The BBC calls for AI companies to strengthen collaboration to improve the accuracy and reliability of information.