You Don't Have to Build Your AI Strategy Alone
July 9, 2026
The excitement around AI is warranted. It has the potential to help your organization synthesize hundreds of reports, surface giving patterns, and uncover insights that would otherwise stay buried in spreadsheets and PDFs.
But that promise comes with questions every social good leader should be asking:
How do we know AI's recommendations are sound?
What happens when technology gets it wrong?
And how do we protect the human judgment and community relationships that make our work effective in the first place?
Here's what we've observed in our work with social good organizations across the country: The ones making real progress with AI aren't those with the biggest technology budgets. They're the ones that have thought carefully about trust — and built the systems necessary to earn and maintain it.
Just as important: They're discovering they don't have to do this work alone.
Lead with principles, not tools
The most common mistake we see is starting with a technology and then figuring out where to apply it. That gets the order backwards.
Before you evaluate any AI tool, get clear on what you can and can't delegate to technology.
Some tasks are genuinely enhanced by AI: synthesizing reports, editing communications, identifying patterns, presenting alternative perspectives and ideas.
Other decisions require human judgment no algorithm should replace: weighing community context, evaluating leadership, making calls on sensitive funding areas.
A growing number of organizations are drawing these lines through an emerging approach known as principles-based computing. Rather than rolling out a tool and bolting on guardrails later, this approach starts with explicit agreements about what the technology should and should not do. Privacy, equity, and human accountability become design principles, not afterthoughts.
Make your AI use visible
Trust erodes quickly when people don't understand how decisions are made.
Internally, disclose how AI is used in your processes — in plain language. If AI-generated analysis informs a recommendation, note it. Externally, be straightforward with grantees and partners about AI's role.
A simple statement in your guidelines — explaining that AI may assist with initial review while humans make all final decisions — goes a long way.
Some leaders worry that openness about AI will invite criticism. In our experience, the opposite is true. Opacity invites suspicion. Transparency builds confidence.
Keep humans at the center
We have a simple principle for AI use: AI proposes, humans decide.
Design every workflow with explicit checkpoints when a person reviews, validates, and takes responsibility for AI-assisted work. AI is a tool. The humans who use it are accountable for the insights they pull from it.
It’s important to make each human interaction a genuine decision point. When something goes wrong — and eventually something will — you want to be able to trace what happened, and then take immediate action to address the issue.
Watch for what’s hidden
AI tools are trained on historical data. They're excellent at finding patterns in what has happened before. They're far less capable of recognizing what's missing.
This matters enormously if you’re centering your work around equity. If your historical patterns underrepresent certain communities, AI trained on that data may quietly perpetuate those biases. It won't flag the gap. It will simply optimize for what it learned.
Human judgment is essential for asking what an algorithm won't: Who isn't in this data? What context is missing?
Build regular reviews into your AI-assisted processes — because AI won't do this work for you. And the better you get at it, the better your AI will get at providing more inclusive data and insights.
The case for building together
Most social good organizations are being asked to develop AI strategy, governance, training, and infrastructure all at once — often with a small team and a limited budget. Our 2025 survey found 75% of community foundations don't yet have a formal AI policy, even as adoption of individual AI tasks runs well above 80%.
That’s an alarming disconnect, and no single organization should have to solve this alone. Increasingly, they don't have to.
We're seeing growing energy across the field for collaborative approaches — organizations co-developing shared principles, pooling their learning, and building infrastructure none of them could build on their own. This is the thinking behind the Community Foundation Alliance Infrastructure Fund, a project we're developing in partnership with the team at empire to help community foundations co-develop industry-specific, principles-based AI together.
Whether through a formal alliance, a peer cohort, or a standing conversation with counterparts at similar organizations, look for opportunities to build alongside others who share your values. The questions of trust, transparency, and equity are too important — and too complex — for every organization to answer in isolation.
The decisions you make — who gets funded, who gets served, how resources flow — have real consequences for real people. AI can help you make better decisions and free up time for the relationship-centered work that matters most. But only if trust is built into every step.
That's the opportunity in front of all of us. It's worth getting right — together.
