Human-Centered AI Governance for Healthcare & Regulated Businesses

AI is already in your business. Is it working for you?

Your team is using AI to draft notes, summarize records, and answer questions, with tools nobody formally approved. Some of that is genuinely helping. Some of it is exposure you would struggle to explain.

I help you tell the difference, then build governance your team will actually follow.

  • Master’s Degree in Computer Science
  • Retired E7, U.S. Air Force
  • Two peer-voted Technical Honors awards at Raytheon
  • Industry Rockstar AI Assisted Consultant Certification, 2026
A lone security architect walking down a secure, windowless data center aisle

Forty-seven years governing systems that had to survive inspection.

Al Goodwin built his career inside classified, air-gapped environments where every decision had to be documented, defended, and still standing a decade later.

He brings that same discipline to businesses with eighty employees instead of eighty thousand, without the enterprise overhead that usually comes attached to it.

0 Years
High-assurance environments, across the U.S. Air Force and Raytheon / RTX
0 Years
Raytheon / RTX, security solutions architecture and governance
0+ Years
A government security inspection score taken from marginal to excellent, then held there
0–0 Clients
Taken on per year, deliberately. A relationship model, not a volume one

It is already in the room. Most businesses have nothing written down about it.

The pattern is the same anywhere the tools arrived faster than the rules. Healthcare is simply where it has been measured.

83%

of medical groups report that AI now plays some role in patient visits. MGMA Stat poll, August 2026

56%

of medical group leaders say they have no AI governance and no formal policy. Another 22% are still working on one. MGMA Stat poll, January 2026, 328 responses

STILL PROPOSED

The biggest HIPAA Security Rule overhaul in two decades has not been finalized. When it lands, most provisions carry roughly a 240-day compliance window

If you handle patient records, client files, financial data, or anything covered by a framework you did not write, you are in the same position. The tools are already inside the business. The governance is not.

HIPAA is the sharpest version of this. It is not the only one. NIST, CMMC and every framework built like them ask the same question: can you show what your AI did, and why.

Adoption arrived first. Governance has not caught up. The rule that will require it is still coming, which makes right now the cheapest time you will ever have to get organized.

Nobody on your team is being reckless. They are being resourceful.

Take a medical practice, because it is the clearest version of the story. The medical assistant got promoted into office manager, scheduler, and billing coordinator. Nobody trained her for three jobs, so she found a tool that helps her keep up. The front desk did the same thing. So did whoever handles follow-up.

Change the job titles and it is the same story in a law firm, an accounting practice, a clinic, or a contractor carrying a compliance obligation. None of them asked permission. None of them were told not to. That is how AI actually arrives in a small business. Not through a purchase order, but through people trying to keep up.

Medical assistant in scrubs juggling scheduling and billing at a clinic workstation
a clinic
Paralegal reviewing client files beside a laptop in a small law office
a law firm
Accountant sorting receipts and client folders next to a laptop
an accounting practice
Contractor in a hard hat checking a tablet outside a job-site trailer in the desert
a contractor carrying a compliance obligation

The gap is not curiosity. It is visibility.

You do not know which tools are in use, what client or patient information has passed through them, what assumptions are shaping the output, or how you would answer if a customer, an insurer, or an auditor asked.

Banning it does not close that gap. It moves the same activity somewhere you definitely cannot see. Ignoring it does not close it either.

What works is making it visible first.

Before you can govern AI, you have to know where it is. Before you can get real value out of it, same thing. Visibility is not the paperwork step before the real work. It is the real work, and everything useful follows from it.

“Your employees aren’t reckless for experimenting with AI. They’re curious. The governance question isn’t how do we stop this. It’s where is it creating value, what data went where, and can we defend it.”
Al Goodwin

What makes this different

I do not sell software. So I have no reason to tell you that you need any.

Most AI advice arrives attached to something for sale. A platform, a license, an implementation contract, a referral fee. That does not automatically make the advice wrong. It does mean you are never quite sure which part is the recommendation and which part is the sale.

I take no tool commissions and I have no product to place. When I tell you a use case is not worth pursuing, that costs me nothing and saves you the money.

A consultant listening carefully as a physician practice owner explains her operation

You do not need another vendor telling you what to buy. You need someone willing to tell you what you actually need.

Assessment before advice.

I map what is actually happening in your business before I recommend a single change. Your biggest problem is often not the one you came in describing.

Still in the work, not retired from it.

I hold a full-time security architecture role at a company that governs systems at national-security scale. What I bring you is current, not remembered.

Enterprise discipline, right-sized delivery.

You get the governance thinking a large organization pays a whole department for, translated into rules your front desk can follow on a Tuesday morning.

Four to five clients a year, deliberately.

This is a relationship model, not a volume one. I want to know your business properly, and I learn as much from you as you get from me. It also means the calendar is genuinely limited.

Both directions, not just the scary one.

Where AI creates value and where it creates risk. Most assessments only go looking for one of those.

The offers

Four ways in. One starting point.

Every one of them opens the same way: finding out what is actually happening.

AI Value & Risk Snapshot

A fast, focused read on where AI is showing up and what it is doing for you and to you.

  • A practical inventory of every AI touchpoint across your operation, including the ones nobody reported
  • A value-and-risk map showing your strongest opportunities and your real exposure, ranked
  • Concrete first moves you can make without freezing momentum

Perfect for: an owner who wants clarity before committing to anything bigger.

From $3,500. 30 days, remote.

Hands mapping sticky notes into an inventory grid beside a laptop

AI Governance Assessment

A close look at where AI touches regulated information, and what has to be true before it spreads further.

  • A full picture of current AI use across your operational and client-facing workflows
  • Gap analysis against the obligations you already carry, including HIPAA and the frameworks built like it, and against the changes the proposed Security Rule would bring
  • A review of the AI features your core systems and vendors have switched on, including the ones you did not choose
  • A written staff AI use policy your team will actually follow
  • Client and patient transparency readiness, so you have a clear answer when someone asks
  • A prioritized roadmap built for a business your size, not for a hospital system

Perfect for: an organization up to roughly 80 employees where staff are already using AI without a written policy.

From $4,500. 30 to 60 days, remote.

Policy binder and stethoscope on a clinic office desk while someone takes notes

AI Value & Readiness Sprint

The full picture. Opportunities mapped, governance gaps closed, and a plan your team can actually execute.

Two meetings is what this costs you in time. Session one is discovery: how the business runs, where AI is showing up, what leadership is worried about. Session two is findings: the value and risk map, the governance gaps, and what to do first. Everything between happens on my side, so your operating schedule stays intact.

  • AI use discovery across every functional area of the business
  • A value-and-risk map covering the whole operation
  • Governance gap analysis against the standards that apply to you, including HIPAA, NIST and CMMC
  • Integration guidance for your highest-value use cases
  • Priority-ranked findings and a 30/60/90-day roadmap
  • An executive-ready Key Findings Report you can hand to a partner or a board

Perfect for: an organization ready to move on AI adoption with something more than enthusiasm behind it.

$7,500 to $9,500. 60 days, two meetings, remote.

Two people arranging sticky notes into a 30/60/90-day roadmap on a whiteboard

AI Governance Roadmap Sprint

Governance built as an operating capability rather than a one-time project.

  • Everything in the Value & Readiness Sprint
  • Governance policy drafting
  • Staff training curriculum designed around your real workflows
  • Vendor and BAA review
  • Documented decision rights, so everyone knows who decides what
  • Quarterly oversight checkpoints and utilization review
  • Ongoing advisory as governance becomes part of how the business runs

Perfect for: an organization making ongoing AI investment decisions that need the architecture to support them.

$15,000 to $20,000. 90 days, remote with periodic on-site.

Leadership team reviewing a printed report around a conference table

Whatever level you start at, the goal is the same: know where AI creates value, understand where it creates exposure, and have a practical way forward.

What I do, and what I do not do.

I assess, advise, guide, and train. I help you decide where AI belongs, how workflows should adapt, and how your people should use it. I do not resell software, take tool commissions, or earn implementation revenue. That is what keeps the advice clean.

When something does need building, you are not handed a report and left to it. I stay involved and bring in trusted partners to execute, so the roadmap turns into something real.

What people say after working with Al.

Doug Meyer portrait
“Having had the pleasure of leading and working with Al Goodwin for many years, I have always found him honest, forthright, and an innovative problem solver. Though to him there are few problems, just opportunities to solve new puzzles. Every new technology is evaluated with a rubric of efficiency, efficacy, cost, level of difficulty, sustainability and compliance. For every effort he is onboard for the duration, supporting the customer and the team. Every team needs an Al Goodwin.”
Doug Meyer, Director of Program Infrastructure Technologies, RTX (Retired)
  • A government security inspection score taken from marginal to excellent, then sustained for over a decade
  • Two peer-voted Technical Honors awards at Raytheon
  • Four months embedded with the IDF in Israel on behalf of Raytheon
  • Contributor, Tucson Business Networking Summer Magazine

You do not just get advice. You get someone willing to stay with the problem until the answer holds up.

The Curiosity Framework

Most AI programs fail because they treat curiosity as the problem.

Some organizations try to stop AI use through restrictions that get quietly ignored. Others rush to adopt without asking whether it fits. Both start from the same mistake. Curiosity is not the problem. It is the raw material.

Assess. Where is AI already being used, and where could it actually help?

Known use and unknown use, where it creates genuine value, and where it is quietly getting in the way. Not a risk scan. A value-and-risk map, and you cannot skip it.

Staff member working across several tools at a clinic workstation
Multi-Model Red Team. Stress-test a tool or workflow from every angle before it goes anywhere near your operations
Standards Anchor. Every recommendation grounded in requirements auditors, insurers and partners already recognize
Governance Shield. Simple rules, clear roles, and review points staff can actually follow
Trust but Verify. Confirm the governance is working while decisions are being made, not in a post-mortem
“Governance is not the thing that slows AI down. Done right, it is what lets responsible teams move faster.”
Al Goodwin

Mission

Governance is not the ceiling. It is the floor you build on.

AI governance written for a Fortune 500 legal department and handed to a forty-person business is not governance. It is paperwork nobody reads. Real governance is practical, visible, and usable by the people doing the work.

The question every owner should be asking before adopting AI is not whether it is safe. It is whether it creates real value, and whether you can govern it well enough to use it with confidence. Those are different questions and they need different answers.

The person running your front office does not need to be replaced by AI. She needs AI running the office so she can go back to the job she was actually hired to do. Right now the strongest retention plan a small business has is teaching its people to use these tools well.

A small professional office building in Tucson at golden hour as a staff member walks to the entrance
“Employee retention is employee training. Teach them AI.”
Al Goodwin

Pull the right levers.

Al Goodwin portrait

Founder story

I got promoted out of the work I loved.

For most of my career I was the person nobody formally appointed but everybody needed. Compliance could not talk to architecture. Architecture could not talk to the system administrators. I sat in the middle and translated, and I was good at it.

Then I got promoted. They took my office. I became the architect, several levels above the room where the actual work happens. I stopped touching systems. I stopped teaching people. I missed it more than I expected to.

Find out what is actually happening, before someone else asks you to explain it.

Fifteen minutes. No pitch, no software, no obligation.

  • A straight read on where your business actually stands
  • The one place your AI exposure is most likely to be sitting
  • What you can fix yourself, and what you cannot
  • No proposal, unless you ask for one

The full engagement runs 60 days and asks two meetings of you. I take four to five clients a year, so if we are not a fit, I will tell you on the call.

Not ready to talk? Get the 2026 AI readiness checklist