Topic

Healthcare AI Compliance

Healthcare AI sits at the intersection of patient safety regulation, data protection law, and AI-specific governance. Here is the layered map and what changes when the AI Act lands on top.

No other industry has more layered AI regulation than healthcare. The good news is that most of the work satisfies multiple regimes at once. The bad news is that nobody can tell you that on the first call.

The compliance footprint depends on what the AI does. Patient-facing clinical AI hits FDA Software as a Medical Device rules in the US and EU MDR or IVDR rules in Europe, with the AI-specific frameworks layered on top. Any system touching protected health information triggers HIPAA, with the privacy rule, security rule, and breach notification obligations that come with it. High-stakes operational decisions (triage, eligibility, prior authorisation) are now also high-risk under the EU AI Act. Most enterprise healthcare AI systems land in at least two of these regimes simultaneously.

The category that gets missed most often is non-clinical operational AI. Revenue cycle automation, denials management, prior authorisation, scheduling. None of this is regulated as a medical device, so it slips through SaMD compliance maps. But it touches PHI, drives decisions that affect patient access and provider finance, and increasingly falls within the AI Act high-risk band. The oversight gap is large, the legal exposure is real, and most organisations cannot tell you who owns it.

The compliance program that works

Three pillars. First, an AI inventory that distinguishes clinical from operational, with each system mapped against every regime that applies. Second, a single shared evidence library so that the SaMD documentation, the HIPAA risk analysis, the AI Act technical file, and the EU MDR technical documentation share data wherever they overlap. Third, one operating process that produces the audit trail every regime separately requires.

How we help

We work alongside your compliance and operations teams, on your own data. Trust, Safety & Compliance covers AI output review and customer and transaction checks. AI Readiness covers strategy, data foundations and hands-on help. We start with a pilot: a small team on real work, nothing goes live until it meets your quality bar, and you get a weekly quality report.

Articles & resources

Frequently asked questions

What regulations apply to AI in healthcare?

It depends on the AI's role. Patient-facing clinical AI usually triggers FDA Software as a Medical Device (SaMD) rules in the US and EU MDR / IVDR rules in Europe. Any AI touching protected health information triggers HIPAA. AI used for high-risk decisions such as triage or eligibility falls under the EU AI Act high-risk category. Most enterprise healthcare AI hits at least two of these regimes.

What is FDA Software as a Medical Device?

SaMD is software intended for medical purposes that performs those purposes without being part of a hardware medical device. The FDA classifies SaMD by risk level and regulates it through the same pathways used for traditional medical devices, with adaptations for the iterative nature of AI systems through frameworks such as the Predetermined Change Control Plan.

How does the EU AI Act interact with EU MDR for clinical AI?

Clinical AI that qualifies as a medical device under EU MDR or IVDR is also a high-risk AI system under the AI Act. The two conformity assessments can be combined under the AI Act's harmonised approach, but the documentation requirements aggregate. Designing for the union of both is more efficient than treating them as separate workstreams.

Where does operational AI fit (the non-clinical kind)?

Operational AI in healthcare (revenue cycle, prior auth, scheduling, denials management) is often missed in compliance maps because it is not regulated as a medical device. But it touches PHI, makes decisions that affect patient access and finance, and increasingly falls under the AI Act high-risk category. The oversight gap here is large and the exposure is real.

Related topics

One body of evidence for every healthcare regime.

Start with a pilot on one AI workflow: a small team, a weekly quality report, and nothing live until it meets your bar.