Medical AI
Physician-authored work on medical AI safety, model evaluation, radiographic output governance and traceability.
-
RheumaView AI Safety Is Now Open
RheumaView AI Safety is open: explore physician-led medical AI safety evaluation, selected research, evidence, and services.
-
Something Different Is Taking Shape at RheumaView
Two new spaces. One clinical perspective. Discover what is taking shape at RheumaView—from clinical AI evaluation to an ambitious author-led resource for patients and professionals. Public access opens within 1–2 weeks.
-
Tested Unicode Masking Did Not Bypass Safeguards, but Unverified Clinical Role Claims Altered Procedural Access: A Two-Stage Evaluation of a Frontier LLM
Explore Olga Goodman’s two-stage LLM safety evaluation of Unicode masking and unverified clinical role claims, presented in a scientific poster.
-
The strongest imaging intelligence is not the one that speaks first. It is the one that can show its work.
Explore why radiographic conclusions should remain connected to the image, location, comparison and uncertainty that support them.
Rheumatology is the art of suspecting inflammation. Radiology is the art of waiting until it finally agrees to be seen.
Rheumatology diagnoses the argument. Radiology waits for the evidence to calcify.