Model Releases

We stress tested many frontier AI models for multimodal medical reasoning (including GPT-5, Claude 3.5, Gemini 2.5 Pro). They’re not ready. …

We stress tested many frontier AI models for multimodal medical reasoning (including GPT-5, Claude 3.5, Gemini 2.5 Pro). They’re not ready. Faulty reasoning, use of inappropriate shortcuts, hallucinat

DGX agentx-post
model-releasesgary-marcus--x

We stress tested many frontier AI models for multimodal medical reasoning (including GPT-5, Claude 3.5, Gemini 2.5 Pro). They’re not ready. Faulty reasoning, use of inappropriate shortcuts, hallucinations. Published today @NatureMedicine https://www.nature.com/articles/s41591-026-04501-8

Source: Gary Marcus (X) | 2026-06-26

Loading related sources…