Four Stages of Nonprofit AI Governance

By Courtney Reeve  ·  May 2026  ·  AI Governance

Someone on your staff used AI this week. Maybe it was ChatGPT to draft a grant narrative. Maybe it was an AI notetaker in a board meeting. Maybe it's embedded in software your team uses every day without anyone flagging it as "AI."

The question facing most nonprofit leaders right now isn't whether AI is in their organization. It is. The question is whether anyone is governing it – and if so, how well.

Across the organizations I work with, a clear pattern has emerged. Nonprofits move through four stages of AI governance, and knowing where your organization sits changes what you need to do next.

Stage 1: Accidental

AI is being used, but no one has defined what's acceptable, what's off-limits, or who's accountable when something goes wrong.

Decisions about AI are being made by individual staff members based on personal judgment – or personal habit. There's no shared framework, no board awareness, and no process for evaluating new tools before they're adopted. Governance, to the extent it exists at all, is happening by default.

What this looks like in practice: A program officer uses ChatGPT to draft donor communications without anyone knowing. An AI notetaker is added to staff meetings without a conversation about data privacy. A new software platform with embedded AI features is approved in a routine budget line without board discussion.

Most organizations are here, or were recently. It doesn't reflect negligence – it reflects how fast AI moved into daily work before governance frameworks existed to receive it.

Stage 2: Intentional

The board starts to engage. AI shows up on the agenda. A policy is drafted, discussed, and often adopted.

This is real progress. Getting AI onto the board agenda requires someone to name the issue, build enough shared understanding for a productive conversation, and navigate the full range of board opinion – from enthusiasm to skepticism to "I don't understand why this is our problem." Adopting a policy is an accomplishment.

What this looks like in practice: The board reviews and approves an AI use policy. The ED communicates it to staff. It lives in the employee handbook or a shared drive.

Where organizations get stuck: Policy adoption feels like governance. It isn't – it's the beginning of governance. A policy that was voted on once and never referenced again isn't being governed. It's been filed. Moving from Stage 2 to Stage 3 requires the policy to change how people actually work.

Stage 3: Operational

Staff practices actually change. People know which tools are sanctioned, what data can and can't be used, and what to do when they're uncertain. The policy isn't just a document – it's a reference point that shapes day-to-day decisions.

This shift is harder than it sounds. It requires the ED to communicate expectations clearly and repeatedly, not just once at rollout. It requires enough clarity in the policy itself that staff can apply it to situations that weren't specifically anticipated. And it requires a feedback loop – a way for staff to surface questions and gaps rather than quietly working around them.

What this looks like in practice: Staff ask before adopting new AI tools rather than after. There's a clear answer to "can I use this?" The ED gets questions about the policy, which means people are actually reading it.

Stage 4: Oversight

The board isn't just an approver. It's an active oversight partner – monitoring AI use, reviewing risk, and receiving regular reporting from the ED.

At this stage, AI is a standing item in the board's governance work, not a one-time agenda topic. The board has enough literacy to ask meaningful questions, evaluate whether risks are being adequately managed, and hold the organization accountable to its own policy. The policy itself gets reviewed periodically, not just filed permanently.

What this looks like in practice: The ED provides a brief AI update at regular board meetings. The board reviewed and updated the policy at the one-year mark. Someone on the board or staff owns the question "is this still working?"

Where Most Organizations Are Right Now

Most nonprofits I'm working with are somewhere between Stage 1 and Stage 2. AI is in the organization. Conversations have started. In some cases, policies exist – but they're incomplete, inconsistently applied, or not yet shaping staff behavior.

Fewer organizations have made it to Stage 3, where practice actually changes. And without that shift, governance hasn't really taken hold regardless of what the policy document says.

Stage 4 is where most organizations aspire to be and few currently are. Not because boards don't care – but because no one has told them what active oversight of AI actually looks like in practice.

That's the gap worth closing. If you want to go deeper on what boards specifically need to do to move from Stage 2 toward Stage 4, the companion piece to this one covers exactly that: What Your Board Actually Needs to Do After Adopting an AI Policy →

Frequently Asked Questions

What are the four stages of nonprofit AI governance?

Nonprofit organizations move through four stages: Accidental (AI is in use but ungoverned); Intentional (deliberate choices have been made about AI use); Operational (governance is embedded in daily practice); and Oversight (the board actively monitors AI as part of its governance role). Most nonprofits are currently in the Accidental or Intentional stage.

How do I know what stage of AI governance my nonprofit is in?

If staff are using AI tools without a formal policy or board awareness, your organization is likely in the Accidental stage. A written AI policy signals Intentional. Routine board-level review of AI use indicates Operational or Oversight-level governance.

What is the difference between AI adoption and AI governance?

AI adoption refers to whether and how AI tools are being used. AI governance refers to the structures, policies, and oversight practices that ensure AI use aligns with the organization's values and risk tolerance. Adoption without governance is where most nonprofits currently find themselves.

How does a nonprofit board move to the next stage of AI governance?

Moving between stages requires deliberate action: adopting a written policy, assigning board-level oversight responsibilities, and building review into existing governance cycles. The AI Governance Policy Suite from Fine Point Consultants provides the tools for each step.

Courtney Reeve is the founder of Fine Point Consultants, a governance and strategy consultancy. She works with nonprofit boards and executive directors on board design, AI governance, and organizational strategy. finepointconsultants.com