FDA-Cleared AI Medical Devices Explained
What FDA clearance means for AI medical devices, how the regulatory pathways work, and what clinicians need to understand before deploying FDA-cleared AI tools.
AI medical devices and software that have received FDA clearance or approval, including 510(k) and De Novo pathways.
The FDA does not approve most medical AI. It clears it, and the difference is not pedantic.
510(k) clearance rests on substantial equivalence to an existing legally marketed device. The agency has not independently tested clinical performance, and the manufacturer was never required to show the device is better than anything. The overwhelming majority of authorized medical AI came through this route.
De Novo authorization covers genuinely novel devices with no predicate, and it creates the classification later devices then cite. Almost every notable first in medical AI went this way. Premarket approval, the route that actually requires clinical evidence of safety and effectiveness, is rare here.
The FDA's public list of AI and machine learning enabled devices passed 950 entries in 2024, and roughly three quarters of it is radiology. That distribution is structural rather than clinical: imaging is already digital, already standardized, and has a mature predicate ecosystem. Pathology and psychiatry authorizations are counted in single digits because those specialties had to build that ground themselves.
It does not tell you the device outperforms a clinician, that it was validated on a population resembling yours, or that performance holds outside the study conditions. Most authorizations rest on retrospective data, which describes how a model scores on stored images rather than what happens to patients when it is switched on.
What FDA clearance means for AI medical devices, how the regulatory pathways work, and what clinicians need to understand before deploying FDA-cleared AI tools.