# Stanford HAI reports medical AI agents finish under half of real clinical tasks

> Original title: Medical AI might be able to ace licensing exams, but on two new benchmarks of real clinical and administrative work, the best agents fini...

- Source: Stanford HAI
- Published: 2026-11-12T17:00:25.000Z
- Bad Influence: https://badinfluence.news/items/lx3mmnab0fna2co7fgn8eam0p
- Original: https://x.com/StanfordHAI/status/2120924688088007095

## Summary

Stanford HAI says two new benchmarks of real clinical and administrative work show the best medical AI agents finished less than half of the tasks. The post contrasts this with the systems' ability to pass licensing exams. Faculty affiliates Nigam and Jon Chen explain the findings in a linked article.

## Post

Medical AI might be able to ace licensing exams, but on two new benchmarks of real clinical and administrative work, the best agents finished less than half of the tasks. HAI Faculty Affiliates @drnigam and @jonc101x explain what they found: https://hai.stanford.edu/news/measuring-what-medical-ai-can-actually-do