Should Your Nonprofit Disclose AI Use to Donors?

The short answer

Yes, in most cases. But the how and when matter more than a blanket yes or no, because different audiences have different concerns and different levels of interest in the details.

The practical question isn't really whether to disclose. It's figuring out which uses of AI are worth mentioning, to whom, and in what context. Using an AI tool to fix a typo in an internal memo is different from using one to draft a grant narrative, and both are different from using one to analyze client outcomes data.

Why disclosure matters in mission-driven work

Mission-driven organizations operate on trust in a way that most businesses don't. Donors give money because they trust the organization to use it responsibly. Clients share personal information because they trust it will be handled with care. Funders award grants because they trust the organization to deliver on its commitments and report honestly about the results.

AI complicates that trust in a specific way: it introduces a question about authenticity. When a donor receives a heartfelt thank-you letter, part of what makes it meaningful is the assumption that someone on staff wrote it because they cared enough to. When a program report lands on a funder's desk, the credibility comes partly from the expertise of the person who wrote it. AI doesn't necessarily diminish the quality of either document, but if the recipient finds out later that AI played a significant role and nobody mentioned it, the trust impact can be real.

Getting ahead of that by being transparent on your own terms is almost always better than being asked about it after the fact.

A framework for deciding what to disclose

I recommend thinking about disclosure across two dimensions: how significant the AI's role was, and how sensitive the audience is to authenticity.

High AI involvement + sensitive audience = disclose proactively. If AI played a major role in creating something that a donor, funder, or client will read and evaluate, mention it. Grant narratives, impact reports, donor appeals, and published content all fall into this category. You don't need a disclaimer on every page, but a brief note in your process is appropriate. Something like: "This report was drafted with AI assistance and reviewed and edited by our program team."

Low AI involvement + any audience = probably not worth mentioning. If someone used an AI tool to check grammar, suggest a subject line, or reorganize their notes before writing something themselves, that's a tool assisting human work. You wouldn't disclose that you used spell check, and this is in the same category.

Any AI involvement + client data = always disclose. If AI tools are processing information about the people your organization serves, disclosure is essential regardless of the context. Clients have a right to know how their information is being used, and in many cases, you may have a legal obligation depending on your state and the type of data involved.

What funders expect

The funder side of this equation is evolving rapidly. Major foundations are increasingly asking about AI use in grant applications, and the trend is clearly moving toward more disclosure expectations rather than fewer.

Most funders who ask about AI aren't trying to catch you doing something wrong. They want to understand how your organization approaches technology adoption, whether you've thought about the ethical implications, and whether you have governance in place. Being able to say "yes, we use AI for specific tasks, here's our policy, and here's how we ensure quality and accuracy" is a much stronger position than either hiding your use or not having a coherent answer.

If your organization uses AI to assist with writing grant applications themselves, it's worth disclosing that directly. Many funders now expect it, and the ones who don't expect it will generally appreciate the honesty. The key is to make clear that the substance and thinking came from your team, and that AI served as a drafting and editing tool rather than the source of your ideas.

How to talk about it with your board

Board members tend to have two concerns about AI use: risk and reputation. The risk concern is about data privacy, accuracy, and liability. The reputation concern is about how it looks if the public or the media learns the organization uses AI.

The most effective way I've seen organizations handle this is to bring the board a clear AI use policy that addresses both concerns directly. When you can show the board that you've thought through which tools are approved, what data is off-limits, and when disclosure is required, the conversation tends to shift from anxiety about AI in general to practical questions about implementation.

It also helps to be specific about where AI saves staff time and what that time gets redirected toward. Board members who understand that AI handles the first draft of a newsletter so that the communications person can spend more time on donor relationships tend to see the value clearly.

What to say to donors

Most individual donors don't need or want a detailed accounting of your internal tools. What they care about is that their gift is being used effectively and that the communications they receive are genuine.

For major donors and long-term supporters who have a personal relationship with your organization, the best approach is usually straightforward: mention it when it comes up naturally, frame it as a way your team works more efficiently, and emphasize that the relationship and the substance remain human-driven.

For broader donor communications like newsletters, annual reports, and appeal letters, a general statement on your website about your approach to technology and AI is usually sufficient. Something brief that explains your organization uses AI tools to support staff productivity while maintaining human oversight of all donor-facing communications. This way, you've been transparent without making it a bigger conversation than it needs to be.

Getting ahead of the conversation

The organizations that handle AI disclosure well tend to have three things in place: a clear internal policy, a short public-facing statement, and staff who are comfortable talking about it when asked. None of these require significant investment, and all three are much easier to create proactively than reactively.

If you don't have a policy yet, that's the place to start. Once you've documented what your organization does and doesn't do with AI, the disclosure question largely answers itself, because you've already thought through which uses are significant enough to merit transparency.