Safety
Biomedical AI may be headed for a replication crisis. (This work below is not about AI-generated reports; it’s about studies of biomedicine …
Biomedical AI may be headed for a replication crisis. (This work below is not about AI-generated reports; it’s about studies of biomedicine that use ML in their methods, and how they are evaluted.) In
Biomedical AI may be headed for a replication crisis. (This work below is not about AI-generated reports; it’s about studies of biomedicine that use ML in their methods, and how they are evaluted.) In a meta-analysis of 210 biomedical AI studies that statistically compared models under cross-validation, 97% used invalid statistical tests. Here's our new preprint https://doi.org/10.64898/2026.05.17.724301 led by @tianchuzeng @kkli20111 @ZShaoshi @ten_photos 1/N
Source: Gary Marcus (X) | 2026-05-21