Adoption Is Outrunning Oversight: Nursing’s AI Governance Gap

Part of Nurse.org’s Nursing AI Watch, our ongoing investigation where we track how artificial intelligence is reshaping nursing and examine who builds it, who governs it, and who’s accountable if and when it fails. Here we look at the gap between how fast AI is being deployed and how slowly the guardrails are being built.

Health systems are racing to deploy AI, but some caution that the guardrails aren’t keeping pace, and perhaps equally as important, aren’t bringing nurses into that governance.

Of the 106 large health systems in the Nursing AI Watch database, at least 26 have a named AI governance body. However, even among the 31 governance structures we could document in detail, only six show a named nurse leader publicly shaping how AI is governed. This piece quantifies that gap, because nurses deserve to know exactly what safeguards are and aren’t in place where they work.

Want to see more Nurse.org articles in your Google results? Add us as a preferred source.

Every one of the 106 large health systems in our research is deploying AI somewhere in care. Far fewer have built a formal way to oversee it.

Our findings show at least 26 of those 106 systems have a named AI governance body, which is a committee or council with a public mandate to review AI tools. That leaves the majority with no oversight body we could confirm at all.

The bigger gap is who actually shapes those committees. Among the 31 governance structures we could document in detail across the 106 systems, only six show a named nurse leader publicly shaping how AI is governed. 

A note on the math: these are floors, not ceilings. Where we could not confirm a governance body, a CNIO, or a disclosure policy from public records, we counted it as not present. The real numbers may be higher. In particular, our count of nurse leaders shaping AI governance is drawn from the 31 governance structures we were able to document in detail — not from all 106 systems — so it, too, is a floor. 

Governance Scorecard: 106 Health Systems Deploying AI

When an AI tool drafts your visit note or flags your chart, do you know? Usually not.

Our findings show at least 16 of 106 systems publish a patient AI-disclosure policy, a public statement that tells patients when and how AI is used in their care. The other 90 either don’t disclose or don’t say where patients can find it.

That matters because disclosure is the front door to trust. A patient who knows AI helped write their note can ask questions, correct errors, and consent with eyes open. Without it, AI runs silently in the background of care, and the people it affects most never get a vote.

The Chief Nursing Informatics Officer, or CNIO, is the role built to bridge nursing and technology. A CNIO understands the bedside and the system, which makes them the natural voice for nursing when AI tools get evaluated.

We could confirm a CNIO at only 12 of 106 systems. That doesn’t mean the other 94 have no informatics leadership; many still do. But a publicly identified CNIO is a signal that nursing has a seat reserved at the technology table, and we could confirm that signal in just over one in ten systems.

Here’s the part that makes the gap urgent. When an AI tool gets something wrong, the nurse is still likely accountable.

The American Nurses Association is direct about this. Its position statement on the ethical use of AI in nursing holds that the nurse remains responsible for their practice and their patients even when the technology fails. The algorithm doesn’t sign the chart. The nurse does.

That’s why a nurse in the governance room isn’t just a courtesy; it’s a necessity if nurses will be the ones ultimately responsible for the technology with their names on it.

We broke down what that accountability looks like in practice in our coverage of the ANA’s AI nursing guardrails. If nurses carry the liability, nurses need a real say in which tools get deployed and how they’re checked.

National guidance points the same way. The Joint Commission and the Coalition for Health AI’s Responsible Use of AI in Healthcare guidance calls for governance structures, monitoring, and clear lines of accountability, not just AI running unsupervised.

>>Listen to The Latest Nurse News Podcast

While governance over AI in healthcare, and within nursing specifically, is still evolving, ideally it should encompass three areas:

• Inclusion: Ensuring nurses are part of the development and decision-making of how and what type of AI will be used. It puts a nurse, ideally a CNIO, at the table from the start, not after launch.
• Disclosure: It discloses AI use to patients in plain language. And it monitors tools after deployment, because an AI that was accurate at go-live can drift.
• Identifying:  Identifying and flagging mistakes is where governance meets safety. Oversight only earns its keep if it catches the failures we documented in our analysis of AI errors, bias, and nurse liability — the scribe mistakes, hallucinated details, and biased outputs that land back on the nurse’s license. Governance is the system that’s supposed to catch those before they reach a patient.

For nurse leaders, the ask is straightforward: push for a named governance body, a nurse seat on it, and a patient-disclosure policy. Unfortunately, most systems we track have none of those three things confirmed.

AI is being deployed faster than it’s being governed. Of 106 large health systems deploying AI:

• At least 26 have a governance body
• 16 disclose AI use to patients
•  12 have a confirmed CNIO (chief nursing informatics officer)

But even among the 31 governance structures we could document in detail, only six show a named nurse leader shaping how AI is governed. The people who carry the accountability are seldom in the room where the decisions get made. Closing that gap starts with a single seat at the table, and our hope is that, by highlighting the issue, we can start changing it at the ground level. 

Related Nursing AI Watch Analysis:

🤔 Does your health system have an AI governance committee — and is a nurse on it? Share you experience in the comments below.

About the data: Nurse.org’s Nursing AI Watch is built from a structured dataset of AI deployments across 106 large health systems, refreshed each reporting cycle. We count only what we can document from the public record — company announcements, peer-reviewed research, news coverage, union and regulatory filings, and official disclosures — and we favor sources independent of the vendor. That means our counts are conservative floors, not exhaustive totals. We update the dataset each reporting cycle and scope every claim to what the evidence supports.

Nurse.org Analysis