Medical superintelligence: foundations and aspirations.
5 affiliations
- 1Microsoft AI, Microsoft, London, UK.
- 2Harvard Medical School, Boston, Massachusetts, USA.
- 3Scripps Health, San Diego, California, USA.
- 4Microsoft AI, Mountain View, California, USA.
- 5Office of the Chief Scientific Officer, Microsoft, Redmond, Washington, USA.
Abstract
Advances in artificial intelligence (AI) have stimulated discussions about medical superintelligence, yet how to define and evaluate this aspiration remains unclear. We discuss how AI technologies can contribute to more effective care, introducing six interdependent dimensions of medical intelligence to frame contributions of AI to healthcare delivery. As AI capabilities and their integration into clinical workflows advance, improving the quality of care will require more than superhuman performance on isolated tasks. It will require systems that harness AI, independently and in collaboration with humans, to surpass today's strongest human-led health systems in integrated medical reasoning across the full arc of care. Realising their value will require understanding how these capabilities complement existing medical intelligence and combining them effectively to improve the quality and accessibility of care.
Related
- The application value of artificial intelligence in medical imaging: foundations, progress, and challenges.
- Sex-specific determinants of leadership aspirations among general internal medicine physicians in Switzerland.
- Clean Hands: From Semmelweis to Surgical Responsibility-Historical Foundations and Contemporary Implications for Infection Control.
- Nutritional Care in Traditional Mongolian Medicine: Historical Foundations and Conceptual Framework Development—A Historical Study
- Transcription Regulatory Sequences in Coronaviruses: Mechanistic Foundations for Antivirals and Vaccines.
BibTeX
@article{liu2026medical,
title = {{Medical superintelligence: foundations and aspirations.}},
author = {Xiaoxuan Liu and Mayank Daswani and Christopher Kelly and Harsha Nori and Nicola Ding and Matthew M Nour and Emma Rocheteau and Craig Sherstan and Viknesh Sounderajah and Adrian D Haimovich and Eric Topol and Mustafa Suleyman and Dominic King and Eric Horvitz},
journal = {BMJ digital health \& AI},
year = {2026},
month = oct,
doi = {10.1136/bmjdh-2026-000395},
url = {https://europepmc.org/article/MED/42920986},
note = {CC BY-NC},
}
RIS
TY - JOUR
TI - Medical superintelligence: foundations and aspirations.
AU - Xiaoxuan Liu
AU - Mayank Daswani
AU - Christopher Kelly
AU - Harsha Nori
AU - Nicola Ding
AU - Matthew M Nour
AU - Emma Rocheteau
AU - Craig Sherstan
AU - Viknesh Sounderajah
AU - Adrian D Haimovich
AU - Eric Topol
AU - Mustafa Suleyman
AU - Dominic King
AU - Eric Horvitz
PY - 2026
DA - 2026/10/22
JO - BMJ digital health & AI
DO - 10.1136/bmjdh-2026-000395
UR - https://europepmc.org/article/MED/42920986
AB - Advances in artificial intelligence (AI) have stimulated discussions about medical superintelligence, yet how to define and evaluate this aspiration remains unclear. We discuss how AI technologies can contribute to more effective care, introducing six interdependent dimensions of medical intelligence to frame contributions of AI to healthcare delivery. As AI capabilities and their integration into clinical workflows advance, improving the quality of care will require more than superhuman performance on isolated tasks. It will require systems that harness AI, independently and in collaboration with humans, to surpass today's strongest human-led health systems in integrated medical reasoning across the full arc of care. Realising their value will require understanding how these capabilities complement existing medical intelligence and combining them effectively to improve the quality and accessibility of care.
SN - 3049-575X
ER -