A public directory of health AI
Does health AI give patients more say, or less?
Most health AI is judged on accuracy and speed. HugoScore judges it on patient agency: whether a tool helps people understand, choose, question, and act. We review the AI patients use directly and the AI used on them behind the scenes, from exam-room scribes to insurance algorithms. Every profile is public, evidence-linked, and open to challenge. Used a tool, or had one used on you? Submit it for review.
Axis scale: 0 = system-run. 100 = patient-chosen. The number shows direction, not product quality. All tools use the same CAIHL-derived criteria.
Public-interest review. HugoScore helps readers ask who controls a health AI tool, what evidence supports it, and what is still unknown.
Draft status: most profiles are AI-assisted public-source drafts. Human review and model provenance are recorded when known; otherwise treat the profile as unverified and open to correction.
Tool profiles
Directory
Draft profiles show public research and the reasoning behind each review. Select a tool to open its full profile. Human review, model provenance, confidence, and limitations are recorded when known.
Intake
Submit a tool or comment
Suggest a tool, challenge a profile, or send evidence that should change a review. Every submission gets a scope decision, and every decision is published in the submission log.
What qualifies: a model makes a judgment or estimate, the output bears on health, and you can name how it changes what a patient understands, decides, or can do. Invisible AI used on patients counts. How scope is decided.