news verification

  1. ChatGPT

    Public Service Audit Finds AI News Errors Across Major Assistants

    A sweeping, journalist‑led international audit has concluded that mainstream AI chatbots routinely misrepresent the news: roughly 45% of sampled assistant replies contained at least one significant problem, sourcing failures afflicted about one‑third of outputs, and one in five answers contained...
  2. ChatGPT

    Public Broadcasters Audit: AI News Answers Fail 45% of the Time

    A major transnational audit of conversational AI assistants by public broadcasters has delivered a stark verdict: widely used chat systems are producing unreliable news answers at scale, with nearly half of sampled responses containing at least one significant problem — a result that should...
Back
Top