Corti vs TORTUS (2026)
The two European clinical AI companies that publish real engineering, and the closest thing this category has to a technical argument between peers. Corti builds healthcare foundation models and sells them as infrastructure, with a sovereign cloud, named models, documented endpoints and a live fact by fact review loop. TORTUS builds an agent that operates the record through its interface using voice, vision and clicks, wraps it in a verification layer that removes unsupported statements before a clinician sees them, and publishes its own error science including a clinical ontology of hallucination and omission. Corti is the better answer if you are building a product or need models under your own roof. TORTUS is the better answer if you are an NHS organisation deploying a finished agent and want the deepest published evidence in this category behind it. Neither is a straightforward United States purchase.
- You are building rather than buying. Three healthcare specialised foundation models stated as trained exclusively on medical data, shipped through documented API endpoints, with sovereign cloud, any cloud or on premise deployment and deliberate portability.
- Setting spread beyond one health system, documented across emergency dispatch, primary care and telehealth, with dedicated language specific models including clinical grade Danish speech to text.
- Named and dated commercial attestations, ISO/IEC 27001 from August 2025 and SOC 2 Type II renewed June 2025, where TORTUS holds NHS assessment frameworks but no named commercial certification located in this pass.
- It publishes error science nobody else in this index publishes: 92.7 percent of detected hallucinations removed as a live metric, a 75 percent reduction in major hallucinations in npj Digital Medicine, and a clinician labelled ontology of hallucination and omission drawn from more than 49,000 transcripts.
- The evidence is prospective, multi site and publicly funded, spanning nine clinical sites and more than 17,000 patients with NHS England backing and independent evaluation, plus a paediatric study at Great Ormond Street and results feeding national policy.
- The agent operates the record through its interface rather than through a built integration, which is a route to coverage that does not depend on a record system vendor agreeing to cooperate, and it picks up patient identity automatically where deployed deeply.
This comparison is published by AI Health Index, an independent research platform that compares healthcare AI vendors objectively. Corti and TORTUS are each graded against the same capability taxonomy, from each vendor's own public materials and the regulatory record, under the AI Health Index verification standard. No vendor pays for placement, and no vendor has reviewed this page. How this evidence is graded
Plain facts
Side by Side
Each record in one paragraph
Written to be quoted whole. Each paragraph states what the AI Health Index verified about the vendor, with the caveats attached. Generated from this pair’s live capability grades, so it moves when a grade moves.
The AI Health Index awards Corti its top capability grade on several axes, including AI Centrality, Autonomy and Oversight Model and Model and Technology Transparency. Set against TORTUS, Corti grades higher on Model Supply Chain Disclosure, Security Certifications and Trust Center and Deployment Model and Data Residency. Grades reflect evidence the AI Health Index could verify at the last review, so a low grade records disclosure the vendor has not published rather than a capability it has been shown to lack.
Source: AI Health Index, August 2026
The AI Health Index awards TORTUS its top capability grade on several axes, including AI Centrality, Autonomy and Oversight Model and Model and Technology Transparency. Set against Corti, TORTUS grades higher on several axes, including Clinical and Operational Evidence, AI Safety and PHI Stewardship and HIPAA and BAA Posture. Grades reflect evidence the AI Health Index could verify at the last review, so a low grade records disclosure the vendor has not published rather than a capability it has been shown to lack.
Source: AI Health Index, August 2026
Questions buyers ask
Should we choose Corti or TORTUS?
On the axes where the AI Health Index separates them, Corti grades higher on Model Supply Chain Disclosure, Security Certifications and Trust Center and Deployment Model and Data Residency, and TORTUS grades higher on several axes, including Clinical and Operational Evidence, AI Safety and PHI Stewardship and HIPAA and BAA Posture. TORTUS leads on the greater share of scored axes, but the split means the decision turns on which constraint is binding rather than on an overall winner.
Where do Corti and TORTUS differ most?
The widest separation the AI Health Index records between Corti and TORTUS is on Clinical and Operational Evidence, where Corti grades C and TORTUS grades A. That axis sits in the AI Capability group, so it should carry the most weight for a buyer whose binding constraint is how much of the work the model itself is trusted to do.
Where do Corti and TORTUS grade the same?
The AI Health Index grades Corti and TORTUS the same on several axes, including AI Centrality, Autonomy and Oversight Model and Model and Technology Transparency. Neither holds an advantage the index can evidence on those axes, so they should not carry weight in a selection between these two.
Related comparisons
Other published head to head assessments involving these vendors or their closest peers. The full set for this category is on the Ambient Scribes page.
Both vendors are European first and neither is a clean United States purchase. TORTUS asserts HIPAA alignment across third party listings with no business associate agreement template or scope statement located, and holds a UKCA Class I registration which is self certified rather than regulator reviewed. Corti's compliance argument is built around European frameworks rather than the United States health privacy rule.
Corti's clinical evidence for its documentation product is thin and partly commissioned market research, in contrast to its strong technical disclosure. Neither publishes a rate card, and neither sells through a self serve path.