Are nonprofits using AI in a way that’s organized, sustainable, and supportive of their teams?

NTEN and The Bridgespan Group‘s recent State of Nonprofit AI Adoption and Governance Report looks at how nonprofit staff and executives are adopting AI, where governance is lagging, and what would help most. Here are four findings we think nonprofit leaders should pay attention to, and what they mean in practice.

1. AI Is Everywhere, but Rarely Integrated

Only 4% of respondents say AI is fully integrated into their organization’s work. Meanwhile, 37% say AI is used informally by individual staff members, without organizational support.

That second number is the one to sit with. Staff using AI on their own isn’t inherently bad. Most people are simply trying to keep up with their workload. But when use is entirely individual and there’s no organizational policy, real problems follow:

• No visibility. Leaders don’t know which tools are in use or what for.
• Data risk. No one has defined what information should never be entered into an AI tool, which matters when you handle donor, client, or program data.
• Inconsistent quality. Everyone prompts differently, reviews differently, and produces different results.
• Duplicated effort. Several people solve the same problem in completely different ways.
• Scattered costs. Individual subscriptions add up without anyone tracking them.

At a minimum, every nonprofit needs a basic AI policy. It doesn’t have to be elaborate. It should answer a few questions:

1. Which AI tools are approved?
2. What information should never be entered into an AI tool?
3. When is human review required?
4. Who approves new tools?

Next, choose the technologies you want your team to use. When one person is on ChatGPT, another on Gemini, and a third using a feature buried in your project management platform, you can’t build shared workflows or train anyone effectively. When everyone uses the same tools, you can create repeatable processes, and ideally a trained AI system that reflects your organization’s voice, standards, and knowledge. The goal isn’t just giving people access to AI. It’s building an organizational approach to it.

2. Embedded AI Is a Great Place to Start

The report found that 76% of respondents are already using AI features embedded in the software they use. That’s encouraging, and it matches our philosophy at Strat Labs: embedded and native features, plus integrations, always come first.

Before you add a new tool, look at what your current platforms can already do. This matters for two reasons:

• Cost. Nonprofit budgets are small. Adding a new subscription every time an AI product launches isn’t sustainable, and many of the platforms you already pay for are adding AI at no extra cost or for a modest upgrade.
• Workflow disruption. Staff already know your existing systems. AI inside a tool they use every day is far easier to adopt than a standalone product that requires a new login, new habits, and new processes.

Our rule of thumb: start with the tools you have, then explore their native AI features, then look at integrations that connect your systems. Only after that should you consider adding something new.

3. Why Nonprofits Are Investing in AI

The top three reasons organizations gave for adopting or investing in AI:

• Reducing staff burden: 56%
• Reducing costs or increasing efficiency: 54%
• Improving consistency of work: 48%

These line up with what we’ve been telling nonprofits for a while. Teams are chronically understaffed and over capacity, working with tight budgets. AI can take on repetitive, time-consuming tasks like first drafts, summaries, and organizing information, so people can spend more time on what requires their judgment, relationships, and expertise.

The third reason is easy to overlook, and it matters a lot in a sector with high turnover. When someone leaves, institutional knowledge often goes with them: the templates, the tone, the process that lived in one person’s head. A well-trained AI model built on your approved language, documented workflows, and organizational context gives new employees a consistent starting point instead of a blank page. It doesn’t replace staff decision-making. It makes the organization less dependent on any one person’s memory.

4. Comfort Levels Will Make or Break Adoption

One of the most telling findings is about comfort. 14% of respondents say they feel very uncomfortable using AI tools as part of their work, 9% feel somewhat uncomfortable, and 15% feel neutral or mixed. That means roughly 38% of respondents aren’t clearly comfortable with AI in their work.

That’s a big deal if you’re trying to roll out official systems. You can have the best tools, policies, and workflows in place, but if a third of your team isn’t on board, adoption will stall.

We’ve seen this repeatedly, and it’s a major part of AI readiness. Readiness isn’t only about your tools and data. It’s about your people. To build comfort and confidence, staff need:

• A clear “why.” What problem is AI meant to solve, and how will it affect their work?
• Role-specific training. Not a one-hour chatbot demo, but practical examples tied to their actual jobs.
• Room to practice. Confidence comes from using the tools, not just hearing about them.
• A safe place for questions and concerns. Hesitation is information, not resistance.

Without that support, even well-designed systems will fail.

From AI Activity to AI Strategy

The report points to a gap between AI use and AI readiness. Nonprofits are already experimenting, and the next step is making that use intentional:

1. Understand how your staff are already using AI.
2. Create a basic AI policy.
3. Start with the embedded features in the tools you already own.
4. Choose a core set of technologies for the whole team.
5. Train and support your people so they feel confident using them.

Download NTEN’s full report for more on governance, staff and executive perspectives, risks, training, and where AI is headed in the sector.

Ready to Build a More Intentional AI Strategy?

We help nonprofits approach AI strategically, with a focus on practical systems, existing technology, staff capacity, and the people who make the work happen. If you’re ready to move from experimenting to building an approach that works for your team, contact Strat Labs today.