Robert Califf

Robert Califf

@robcaliff.bsky.social

1K followers62 following678 posts

My answer to Wachter's list at the end is "all of the above". We need transparency about who controls the purpose, someting like FSMA rules for food safety applied to AI (what are the values?) and local, recurrent validation because an AI is like an ICU patient-needs constant monitoring.

If you know someone who wants to go deep with clinical research on the global stage, this is it.

Curious about a career in clinical research? Join the DCRI Clinical Research Fellowship Program webinar Oct. 22, 11 a.m.–12 p.m. ET. Open to residents, fellows, faculty & medical students. Free to register: duke.zoom.us/webinar/regi... Learn more: dcri.org/fellows #ClinicalResearch #Fellowship

robcaliff272993.substack.comWorking Together on Truth: The Story of Coethia, Part IIToward a Better Understanding of Research and Evidence in Health and Healthcare
Robert Califf @robcaliff.bsky.social

Reposting a substack on the rationale for Coethia, an organization attempting to understand how to produce reliable health information on social media and how to improve communication with individuals in the clinic and public health system. robcaliff272993.substack.com/p/working-to...

but for sure the direction of AI prowess challenges clinical educators and professional societies to define the skills of clinicians that take advantage of AI and retain humanism.