AI chatbots provide dangerous levels of medical misinformation🚨
Don't get your advice from AI.
Generative artificial intelligence-driven chatbots and medical misinformation: an accuracy, referencing and readability audit
Study Details
This new study investigated responses of 5 popular AI chatbots…
👉 Gemini
👉 DeepSeek
👉 Meta AI
👉 ChatGPT
👉 Grok
Chatbots were prompted with 10 questions from five categories (cancer, vaccines, stem cells, nutrition and athletic performance) and were assessed for scientific accuracy 🔍
Key Findings
👉 49.6% of responses were problematic
👉 30% were somewhat problematic
👉 19.6% were highly problematic
Response quality did not differ significantly among chatbots.
…BUT Grok generated significantly more highly problematic responses than would be expected under a random distribution ⚠️
💉 Performance was strongest in vaccines and cancer questions
❌ Performance was worst in stem cells, athletic performance and nutrition
Conclusion
AI is highly prone to providing medical misinformation, especially in the athletic performance and nutrition domain‼️
If you want reliable, trustworthy and evidence-based information, ask a qualified and experienced practitioner ✅
Reference
https://pubmed.ncbi.nlm.nih.gov/41980854/
Performance Nutrition Digest is powered by Healthspan Elite ⚡️
Healthspan Elite is a science-led sports nutrition range developed in collaboration with leading experts to support athletic performance, recovery, and health.
Use code: TCDIGEST25 for 25% off at https://www.healthspanelite.co.uk/ 💪





