G4E v0.9 · Last reviewed September 1, 2026
Where G4E comes from.
I built G4E because I kept having the same conversation. A health system has an AI committee, a pilot or two, a policy somebody wrote last year, and no clear line between any of it. The pieces exist and nothing connects them.
G4E is how I organize that problem. Five capabilities: Govern, Explore, Educate, Enable, Empower. It is a way to see the whole thing at once and decide what to work on next.
What it is built on
Fifteen years in healthcare IT, including running analytics and enterprise platforms at El Camino Health. More than twenty-five roundtable sessions with healthcare IT and AI leaders talking through what is happening inside their organizations. The AIGP certification in AI governance. Current work building an AI literacy program for a multi-hospital health system. A fractional COO seat at a healthcare AI company, which is a different view of the same problem.
Where each pillar stands
I have not run a complete G4E program end to end at one organization. This field has not existed long enough for many people to say they have, and I would want to see the receipts from anyone who claimed it. So here is the honest state of each pillar.
- EducateIn delivery
- Being built and delivered now for a multi-hospital health system. This is the most tested part of the model.
- GovernUsed in pieces
- Drawn from published frameworks, the roundtable series, and governance work with clients. I have not yet run it as a full program at one organization.
- EnablePartly built
- Built from role transformation research. The physician documentation arc is complete. Other roles are in development and marked that way on the page.
- ExploreAssessment working, pillar untested
- The readiness assessment works and has been used with organizations. The wider Explore practice around it is structured and untested.
- EmpowerNewest
- The research is published and sourced. The measurement practice behind it is still a hypothesis.
This will change
G4E sits at v0.9 and the version number is honest. As I run more of it, parts will get sharper and some parts will turn out to be wrong. When that happens I will change them here and date the change rather than quietly leaving the old version up.
Tell me what is wrong
If you read something here that does not match what you have seen inside a health system, I want to hear it. That goes double for the glossary and the materials, which cover a lot of ground and are the places I most expect to be corrected. Email me at hello@henecorp.ai. I read everything and I will fix what needs fixing.