AI chatbots still can’t accurately answer high-level history questions: study
Researchers found that LLMs often extrapolate from prominent historical data but struggle with more obscure details. For ...