Health AI

Health AI is an American health technology company that develops AI governance tools and evidence-based ingredient safety databases. Its public positioning is built around a single discipline it calls "evidence discipline": retrieve the relevant record, preserve provenance, grade what the source can support, and state the result no more strongly than the evidence allows. The company explicitly separates this discipline from any shared database or risk model, noting that its products do not share one database or one risk model but do share that rule. Health AI's consumer-facing product is Clarity, described as an on-demand ingredient safety advisor. Clarity checks products against ingredient records, citations, and condition-specific evidence tiers, then uses AI to interpret the result in plain language. It is designed to keep skincare, food, and supplements in one place, with patterns tracked over time. The company states that Clarity covers 4M+ products scannable by barcode, photo, or search, and 29K+ evidence-graded ingredient records across food, supplement, skincare, and label contexts, including PMID-backed citations. It applies 32 decision dimensions across allergens, histamine, rosacea, HS, pregnancy, fertility, asthma, dose, and route, and shows evidence tier and confidence where a result needs judgment. Clarity is positioned for pregnancy, breastfeeding, histamine/MCAS, rosacea, HS, allergy/asthma, and fertility/TTC audiences, and the company emphasizes that it offers no brand-paid verdicts and is condition-aware and evidence-tiered. Its stated knowledge sources include NIH PubMed, LactMed, FDA, EFSA, SIGHI, and DSLD. A second product, Constat, is described as having distinct users and evidence domains from Clarity, with clear product boundaries. Health AI also offers RIGOR, a framework that connects requirements, implementation, governance, operational proof, and runtime monitoring across five domains. The company states that machine access is provided through two MCP endpoints, which it describes as a channel into each product's evidence rather than a third product or a separate promise. Beyond products, Health AI offers AI literacy for institutions and AI consulting for decision-makers. Health AI publishes research on AI governance, including a revised report arguing that AI governance can be a revenue strategy. The company says the report separates what primary sources establish from what remains a commercial thesis, maps the authorization-to-value gap, defines evidence architecture, and explains how RIGOR and Constat fit together without claiming that governance alone causes revenue. It also references a 7-page PDF edition and a whitepaper on provisional approval and the evidence Health AI must keep. The company cites a published field accuracy figure of 90.1% for a dual-VLM inspection pipeline across 203 vehicles with zero failures and a DOI-archived method, presented as an example of its evidence-grading approach.

Health AI is an American health technology company developing AI governance tools and evidence-based ingredient safety databases. Public information about the company's founding, founders, and corporate history is limited; the available evidence comes primarily from its official website and a Wikidata entry describing it as an American health technology company. The company's product architecture centers on a stated principle it calls "evidence discipline." According to its website, different systems share the same discipline: retrieve the relevant record, preserve provenance, grade what the source can support, and state the result no more strongly than the evidence allows. Health AI explicitly notes that its products do not share one database or one risk model, but they share this rule. This framing appears to be the company's core differentiator as it markets to both consumers and institutions. Health AI's consumer-facing product, Clarity, is described as an evidence-based ingredient scanner and checker. It checks products against ingredient records, citations, and condition-specific evidence tiers, then uses AI to interpret the result in plain language. The company states that Clarity covers 4M+ products scannable by barcode, photo, or search, and 29K+ evidence-graded ingredient records across food, supplement, skincare, and label contexts, including PMID-backed citations. It applies 32 decision dimensions across allergens, histamine, rosacea, HS, pregnancy, fertility, asthma, dose, and route. Clarity is positioned for pregnancy, breastfeeding, histamine/MCAS, rosacea, HS, allergy/asthma, and fertility/TTC audiences. The company emphasizes that Clarity offers no brand-paid verdicts and is condition-aware and evidence-tiered. Its stated knowledge sources include NIH PubMed, LactMed, FDA, EFSA, SIGHI, and DSLD. A second product, Constat, is described as having distinct users and evidence domains from Clarity, with clear product boundaries. Health AI also offers RIGOR, a framework that connects requirements, implementation, governance, operational proof, and runtime monitoring across five domains. The company states that machine access is provided through two MCP endpoints, which it describes as a channel into each product's evidence rather than a third product or a separate promise. Beyond products, Health AI offers AI literacy for institutions and AI consulting for decision-makers. Health AI publishes research on AI governance. Its website references a revised report arguing that AI governance can be a revenue strategy, stating that the report separates what primary sources establish from what remains a commercial thesis, maps the authorization-to-value gap, defines evidence architecture, and explains how RIGOR and Constat fit together without claiming that governance alone causes revenue. The company also references a 7-page PDF edition and a whitepaper on provisional approval and the evidence Health AI must keep. It cites a published field accuracy figure of 90.1% for a dual-VLM inspection pipeline across 203 vehicles with zero failures and a DOI-archived method, presented as an example of its evidence-grading approach.

Evidence-first health technology: knowledge before AI, provenance-preserving, evidence-tiered, and condition-aware, with no brand-paid verdicts.

• Clarity

  On-demand ingredient safety advisor that checks products against ingredient records, citations, and condition-specific evidence tiers, then uses AI to interpret results in plain language. Covers 4M+ products and 29K+ evidence-graded ingredient records across 32 decision dimensions.

• Constat

  A product with distinct users and evidence domains from Clarity, with clear product boundaries.

• RIGOR

  A framework connecting requirements, implementation, governance, operational proof, and runtime monitoring across five domains.

1. 1Health AI | Clarity, Constat and Evidence Tools for AI Agentshealthai.com
2. 2Clarity | Evidence-Based Ingredient Scanner and Checkerhealthai.com
3. 3Health AIwikidata.org