Qualified Health

A two-year-old company already runs inside hospitals you've heard of

Walk into a health system using Qualified Health and you will not see the company at all. You will see a clinician finishing a note faster, a billing team closing a gap, an AI assistant answering a question with a citation and a timestamp. The product is invisible on purpose. It is the layer underneath - the rails the AI rides on, the place where every model's decision is logged, governed, and measured.

That is the strange thing about Qualified Health. It sells the part of artificial intelligence that nobody demos at a conference. No magical chatbot. No avatar. Just the operating layer that decides whether a hospital's AI is allowed to do anything at all. Roughly 7% of U.S. hospital revenue now flows through systems that have plugged into it, and more than half a million people use it. The company is two years old.

Hospitals had pilots. What they did not have was control.

By 2023, every health system in America had an AI pilot. Or twelve. A scribe here, a triage tool there, a vendor demo that wowed the board and then quietly never reached a patient. The technology worked. The governance did not. Each tool arrived with its own data connection, its own dashboard, its own promises, and its own blind spots.

The result was a familiar healthcare paradox: enormous activity, very little scale. A hospital could run forty AI experiments and still not be able to answer the only questions that matter to a Chief Medical Officer - is this safe, who is accountable, and did it actually help anyone? Outsourced point solutions made it worse. Hand the roadmap to a vendor and you no longer own your own future.

Qualified Health's bet is almost rude in its simplicity: the bottleneck in healthcare AI is not intelligence. It is trust. And trust, inconveniently, is not a feature you can bolt on at the end. It has to be the floor.

From self-driving cars to self-governing hospitals

Justin Norden has an MD, an MBA, and an MPhil, which is the academic equivalent of refusing to pick a lane. Before Qualified Health he built Trustworthy AI - a company about exactly what the name says - and sold it to Waymo, Google's self-driving division. He then spent time as a venture partner backing healthcare AI startups, which gave him an unusually clear view of why most of them stalled.

He did not build Qualified Health alone. He recruited people who had spent their careers inside the systems the company now serves: a former president of the Institute for Healthcare Improvement, an operator who helped build Haven and Evolent, and a former VP of AI from one of the country's largest insurers. The pattern is deliberate - this is a company of insiders, not disruptors shouting from the outside.

Four founders, zero outsiders. The kind of bench you assemble when your plan is to be trusted rather than admired.

Integrate once. Govern everything.

The platform's core promise is one connection. Instead of wiring each new AI tool into Epic and the billing stack and the operational systems separately, a health system integrates its data once. On top of that single layer sits a library of pre-validated solutions for high-impact workflows, plus builder tools for teams that want to make their own agents.

What makes it a platform rather than a toolbox is the governance running underneath all of it - real-time oversight, audit trails, role-based access, post-deployment monitoring. Every AI action is watched, logged, and measured against clinical, operational, and financial outcomes. It is, in spirit, the compliance department finally given software that can keep up with the engineers.

The receipts, because skeptics in healthcare are correct to demand them

Healthcare has buried more AI promises than any other industry, so a claim without a number is just a press release. Here are the numbers Qualified Health and its backers put on the table. At the University of Texas Medical Branch, the company reports more than $15M in run-rate impact within six months - achieved by integrating data systems, deploying AI assistants, and automating workflows rather than by any single headline feature.

Numbers the company markets; the "~7%" figure refers to revenue of the systems it serves, not market share of AI spend.

The customer list reads like a tour of American academic medicine, and the investor list reads like a vote of confidence from people who have seen plenty of healthcare AI fail. NEA led the Series B. Anthropic - through Menlo's Anthology Fund - showed up too, which is notable for a company whose entire pitch is governing the very models Anthropic builds.

Health systems publicly named as working with Qualified Health.

A company legally obligated to mean it

Qualified Health is a public benefit company, which means its charter requires it to weigh mission alongside profit. In an industry where "AI for good" is usually a slide, that is a structural commitment rather than a slogan. The mission is plain: enable safe, scalable AI across healthcare through unified governance and trusted infrastructure - and let health systems own their own roadmap instead of renting it.

It is a quieter ambition than most AI companies advertise. There is no talk of replacing doctors, no countdown to superintelligence. The vision is closer to a utility: every hospital able to build, deploy, and govern its own AI safely, at scale, with the receipts to prove it worked.

Why the invisible layer is the one that matters

Go back to that hospital. The clinician finishing the note faster, the AI answer with a citation, the billing gap quietly closed. None of it looks like a revolution, and that is the point. The most consequential infrastructure is the kind you stop noticing - electricity, plumbing, the EHR everyone complains about but no one would remove.

If healthcare AI is going to scale past the pilot stage, something has to sit underneath it and answer for it. Qualified Health's bet is that this layer is not a commodity but the whole game - that the company governing the models will matter more than any single model. Two years in, with $155M and a meaningful slice of American hospital revenue running through it, that bet looks less like a hunch and more like a head start.

The chatbot was never the hard part. The trust was. Qualified Health decided to sell the trust.