From governance and risk considerations to emerging best practices, I asked nonprofit finance leaders to weigh in on their experiences applying AI within their organizations. GHJ’s recent nonprofit roundtable on technology and AI revealed goals and pressures among organizational leaders when it came to AI. Among the group, nonprofit finance professionals were focusing efforts on improving efficiencies, increasing analytical capacity, supporting compliance and helping teams “do more with less” without compromising mission delivery. Speaking with these finance professionals about AI illustrated that many organizations are entering this work at very different stages, and the pace of change itself makes experimentation more important than perfection. 

Nonprofit finance leaders starting to utilize AI found these key takeaways to be the most important for their organizations right now: 

START WITH THE BUSINESS PROBLEM, NOT THE TECHNOLOGY

The most successful use cases begin by identifying a specific challenge or inefficiency. Rather than asking, “Where can we use AI?”, organizations can ask, “What is slowing us down, and can AI help solve it?” Several of the nonprofit leaders also stated that they are looking to adopt AI as a form of additional capacity to help small teams handle increasing workload demands. 

Practical takeaway: Organizations should start by understanding the problem they are trying to solve, determine what capabilities are required and then select the most appropriate technology. Tool selection should always follow strategy, not drive it.

AI IS MOST EFFECTIVE WHEN IT SITS ON TOP OF STRONG FOUNDATIONS

Across the organizations I spoke with, most are already experimenting with AI in some capacity, whether formally or informally. A number of these nonprofits are also integrating AI into operational systems like NetSuite, ClickUp or Blackbaud environments, while others remain in early exploration phases. They described using AI to summarize board finance packets, draft grant narratives, reconcile policy documentation and accelerate internal research while maintaining human review.

Practical takeaway: To get started, AI should be viewed as the top layer of a system built on good data, connected systems, sound processes and clear business workflows. AI can accelerate analysis and automate tasks, but the quality of outputs is directly tied to the quality of inputs. Organizations that struggle with data quality or fragmented systems will likely not realize the full value of AI until processes and data improvements are addressed.

GOVERNANCE SHOULD EVOLVE ALONGSIDE EXPERIMENTATION

One of the clearest findings from speaking with these leaders was that governance maturity is still developing across the sector. Some organizations have formal AI policies or steering committees in place, and others are actively drafting policies. Several finance leaders candidly acknowledged they currently have no formal governance structure and recognize that as a risk area.

Practical takeaway: Overall, these nonprofit leaders found that organizations do not need fully developed governance frameworks before beginning to explore AI. However, policies, guardrails and oversight should develop in parallel with experimentation, not after AI adoption becomes widespread. An AI Safe Usage Policy is a practical starting point.

AI LITERACY MATTERS MORE THAN CHOOSING THE “BEST” TOOL

Given the pace of change, the leaders agreed that organizations should focus on building curiosity, experimentation and AI literacy rather than trying to identify a single winning platform. Today's leading model may not be tomorrow's leader when it comes to this technology. Steps to take to initiate organizational AI literacy can include: 

Champions and Super Users Help Drive Adoption

The group found that having internal champions in place can play an important role in encouraging experimentation, sharing lessons learned and helping colleagues build confidence with new tools. Several organizations found adoption accelerated after designating finance staff as AI champions who documented successful prompts and shared approved use cases during monthly team meetings.

Define the Role of the AI Tool

Define the role each AI application is expected to play. Whether acting as a research assistant, first-draft writer or analytical support tool, establishing clear boundaries improves consistency and helps users understand when human review is required.

Human Judgment Remains Essential

Across all use cases discussed, the group emphasized that AI can enhance productivity and decision support, but it does not replace critical thinking, professional judgment, ethical considerations or accountability.

As AI capabilities expand, the role of staff will evolve from performing individual tasks to directing, reviewing, coordinating and stewarding work performed by intelligent systems. Success will increasingly depend on judgment, oversight and the ability to orchestrate resources effectively.

WHAT THESE TAKEAWAYS TEACH 

The nonprofit sector is still in the early stages of AI adoption. Organizations do not need to wait for perfect policies or perfect technology before beginning, but they do need thoughtful governance, reliable data and clear accountability. Those foundations will determine whether AI becomes a meaningful productivity tool or simply another technology experiment.

Several organizations shared that developing acceptable-use policies, evaluating AI vendors and establishing governance frameworks were among the first areas where external advisors added significant value to their nonprofits. To learn more about how GHJ supports nonprofit CFOs with AI readiness, reach out to the Nonprofit Practice.

This article was inspired by a recent CFO roundtable GHJ held with finance leaders at nonprofit organizations.