Loading blueprint versions...
Please wait while we gather all the unique runs for this blueprint.
Please wait while we gather all the unique runs for this blueprint.
Tests whether models understand invisible disabilities (conditions not immediately apparent) and episodic/fluctuating disabilities (conditions that vary day-to-day).
Key finding: Up to 80% of disabilities are invisible. Conditions like ME/CFS, fibromyalgia, POTS, and mental health conditions face documented histories of dismissal and skepticism that may be reflected in training data.