An independent testing lab for clinical AI. Products run inside a production-grade simulated hospital, complete with EHR, radiology, lab, telehealth, and patient portal, populated entirely with synthetic patients. Licensed clinicians work blinded against your product in live scenarios. You leave with a safety scorecard across five domains, before the product ever touches a real patient.

Three groups: clinical AI developers, healthcare deployers and research and development teams; developers who need independent, reproducible evidence of clinical safety before a first hospital pilot. And health systems and academic groups who need neutral third-party validation of a vendor’s claims, for procurement due diligence and for readiness against Joint Commission RUAIH and URAC.

Sandboxes tell you whether your code runs. We tell you whether a clinician trusts the output, and exactly where it breaks. We use Synthea ourselves to generate the patient population. The difference is what sits on top of it: a longitudinal EHR, an adversarial failure layer, and blinded clinical review. Complementary, not competing.

No. Tap Hospital performs testing and validation. We report what we found, how we found it, and what it means, in a scorecard and a written explanation across five domains. We do not issue a seal of approval and we are not a certification body. The distinction matters. A certification asks you to trust the issuer. Our report lets you inspect the method and the evidence yourself, and tells you which scenarios broke the product, so you can act on it.

Worth stating plainly, our testing is not a step outside the accreditation path. It is the evidence that path runs on. Certification schemes and hospital AI governance committees increasingly ask for independent testing performed by someone other than the vendor and buyer, and that is exactly what a Tap Hospital engagement produces.