Health AI · founded by Olga Lavinda, PhD

Evidence tools for people and AI agents.

Health AI is the company behind Clarity and Constat. RIGOR is the evidence method. MCP Queen adds server discovery, operational evidence, and reviewed field reports. Use the systems in a browser or connect them to compatible Claude and OpenAI workflows.

2 live products · browser + MCP access · primary-source citations · public methods
One evidence discipline · proof, not adjectives

Different systems. The same discipline.

The products do not share one database or one risk model. They share a rule: retrieve the relevant record, preserve provenance, grade what the source can support, and state the result no more strongly than the evidence allows.

claimretinol · pregnancy
01 · retrievethe recordLactMed · DermNet
02 · gradetier + stateGold · cited
03 · interpretthe verdictplain language
AVOID◈ Gold · cited · LactMed
90.1%
Published field accuracy of the dual-VLM inspection pipeline: 203 vehicles, zero failures, DOI-archived method.
5 domains
RIGOR connects requirements, implementation, governance, operational proof, and runtime monitoring.
2 products
Constat and Clarity have distinct users and evidence domains, with clear product boundaries.
2 MCP endpoints
Machine access is a channel into each product's evidence—not a third product or a separate promise.
Research

Why AI governance is a revenue strategy.

Our revised report separates what the primary sources establish from what remains a commercial thesis. It maps the authorization-to-value gap, defines evidence architecture, and explains how RIGOR and Constat fit together without claiming that governance alone causes revenue.

New Insight: Provisional Approval and the Evidence Health AI Must Keep →

Start anywhere

One company. Clear roles.