eWeek

Grok Had Highest Share of ‘Highly Problematic’ Health Answers in BMJ Open Audit

BMJ Open found 49.6% of AI chatbot health answers were problematic. Grok performed worst, raising fresh concerns over medical ...