Most nonprofit jobs are three jobs. The development director also runs events and the newsletter; the program manager also writes grant reports and the board dashboard. AI helps most with the writing that never ends: proposals, thank-you letters, impact stories, meeting summaries, and the volunteer email that goes out every Thursday.
Protect two things: trust and truth. Donor records, client information, and beneficiary stories are the most sensitive data you hold and do not belong in a consumer chatbot. A proposal with an invented statistic or a made-up participant quote can cost you a funder for good. AI drafts; you verify against your financials, your program data, and real people who consented to be quoted.
The upside is real. A small team with good prompts and a few reusable templates can produce work that used to take a bigger staff, and can answer more funding opportunities without burning out. Start with the writing, keep the data clean, and build from there.
Quick wins this week
- Paste a funder's RFP and your last successful proposal (with client details removed) and ask for a gap analysis: what the funder asks for that your standard language does not cover.
- Draft five versions of a thank-you letter for different gift levels, then personalize each one by hand with details only you know.
- Summarize a two-hour board meeting recording into decisions, action items, and open questions, and send it out the same day.
- Ask for a first draft of your Form 990 Part III program service descriptions from your annual report language, then have your accountant review it.
What AI can do for nonprofit professionals, task by task
Grant proposal drafting and RFP response
Give the assistant the RFP, the funder's stated priorities, your program description, and past proposal language with client details removed. Ask for a draft that answers each question in order, in the funder's vocabulary, within their word limits. Verify every number against your audited financials and program data, and check whether the funder has an AI-use policy for applicants.
Donor communications: appeals, thank-yous, and stewardship
Describe the campaign, the segment, the ask, and the story you want to tell, and request drafts at different lengths for different channels. Never paste donor names, giving histories, or contact details into a consumer tool; segment in your CRM and merge there. Read the draft for anything generic, because donors notice.
Impact reports and grant reporting
Paste aggregated outcome data (no client identifiers), the funder's reporting template, and last year's report, and ask for a draft that ties each outcome to what you promised in the proposal. Ask it to list every figure it used, check each against the source, and make sure any story comes from a real participant with consent on file.
Board and leadership materials
Give the assistant your agenda, notes, and financial summary and ask for a board memo, a dashboard narrative, or a consent-agenda summary in your usual format. Meeting assistants can transcribe board meetings and draft minutes; check your bylaws and any open-meeting rules that apply, and have the secretary review before approval.
Research on funders and the field
Use Perplexity or a chat assistant with web access to research foundations, their recent grants, and their priorities, then confirm on the funder's own site and in Candid before you act. For individual prospects, respect the line between public information and profiling, and follow your donor privacy policy.
Automating the small stuff
Connect your form tool, CRM, and email with Zapier or Make so a new donation queues a thank-you draft for review, a volunteer application creates a task, and event registrations land in a sheet. Keep a person in the loop for anything donor-facing, and test each automation with dummy data before it touches real records.
Prompts for nonprofit professionals
Replace the bracketed placeholders, paste into any chat assistant, and iterate on the result.
Grant proposal section from an RFP
Act as an experienced nonprofit grant writer. Using the RFP and our materials below, draft the [SECTION NAME] section in [WORD LIMIT] words or fewer. Answer exactly what the funder asks, in their order, using their vocabulary for outcomes and populations. Do not add statistics or claims that are not in our materials; where a number would strengthen the case, write [NEED DATA] and tell me what to find. End with a checklist of every fact and figure you used. RFP: [PASTE RFP TEXT] Funder priorities: [PASTE OR DESCRIBE] Our materials, client details removed: [PASTE MATERIALS]
Tip: The [NEED DATA] marker keeps the model from inventing a statistic. Fill each one from your real records.
Donor thank-you letters by gift level
You are a development director known for warm, specific donor letters. Write thank-you letters for these gift levels: [GIFT LEVELS]. Organization: [ORGANIZATION NAME AND MISSION IN ONE SENTENCE]. What the gift makes possible this year: [ONE OR TWO CONCRETE PROGRAM FACTS]. Tone: [TONE, FOR EXAMPLE WARM AND PLAIN]. Each letter under 180 words, opening with the impact rather than the organization, including one detail a donor could picture, and leaving a marked spot for a personal line I will add by hand. No donor names; I will merge those in our CRM.
Tip: Personalize by hand. The letter the model writes is the floor; the line you add is why the donor gives again.
Impact stories from program data
Act as a nonprofit communications writer. Turn the aggregated outcome data below into three 120-word impact stories for [AUDIENCE, FOR EXAMPLE NEWSLETTER OR ANNUAL REPORT]. Each story leads with a concrete change, uses one number, and ends with what comes next. Do not invent any individual, quote, or anecdote; where a human example belongs, write [REAL PARTICIPANT STORY WITH CONSENT] and describe what kind of story would fit. List every number you used at the end. Program: [PROGRAM NAME AND WHAT IT DOES] Data: [PASTE AGGREGATED OUTCOMES, NO CLIENT IDENTIFIERS]
Tip: Never let the model write a beneficiary. Real stories, real consent, every time.
Board meeting summary from a transcript
You are an executive assistant to a nonprofit executive director. From the transcript below, produce: (1) decisions made, each with the motion, who moved and seconded if stated, and the vote; (2) action items with owner and due date; (3) open questions; (4) a 100-word summary for absent board members. Use only what is in the transcript; if an owner or date is not stated, write 'not stated'. Include nothing from executive session.
Transcript: [PASTE TRANSCRIPT]Tip: Check your bylaws on what minutes must contain, and have the board secretary review before distribution.
Funder research with sources
Act as a prospect researcher for a [ORGANIZATION TYPE] in [LOCATION] working on [PROGRAM AREA]. Find foundations and corporate giving programs that funded similar work in the last three years. For each: name, stated priorities, typical grant range if published, application process, deadline if known, and a link to the source. Mark anything you could not confirm from a primary source as 'unverified'. Rank the top five by fit with a one-line reason each. Ask me clarifying questions first if the program area or geography is unclear.
Tip: Confirm every funder on their own website and in Candid before you spend an hour on an application.
Want a prompt for something else? Use the Prompt Builder or browse brainstorming and strategy prompts, finance and analysis prompts, leadership and communication prompts, presentations and reports prompts.
Skills to build
Keeping donor and client data out of the prompt
Why: Your donor database and client files are the organization's most sensitive assets. One paste into the wrong tool can breach donor trust, state privacy law, or a funder contract.
How: Work with aggregates and descriptions. Merge names in your CRM, never in the chat tool. Write a one-page rule for staff and volunteers about what may and may not go into AI tools.
Verifying every number and story
Why: Funders check. A proposal or report with an invented statistic or a mismatch against your Form 990 damages credibility you cannot easily rebuild.
How: Ask every prompt to list the figures it used. Tie each one to your financials or program data before submission. Use only real participant stories with documented consent.
Building a reusable materials library
Why: Most proposals reuse the same organizational history, program descriptions, and outcomes. A clean, current library makes every draft faster and more accurate.
How: Keep a folder of approved boilerplate, current outcome figures with dates, and your best past proposals. Load it into a Claude project or a custom GPT so every draft starts from approved language.
Writing in the funder's language
Why: Reviewers score against their own criteria and vocabulary. Matching their terms for outcomes and populations makes a proposal easier to score well.
How: Paste the RFP and ask the assistant to extract the funder's key terms and criteria before drafting. Check the finished draft against that list.
Simple automation with a person in the loop
Why: Small teams lose hours to manual data entry between forms, the CRM, and email. Automation returns those hours if a person still approves anything donor-facing.
How: Pick one recurring handoff (donation received to thank-you queued), build it in Zapier or Make with test data, and add a review step before anything sends.
Tools worth knowing
ChatGPT
The general-purpose AI assistant most of your coworkers already use.
Claude
A careful writing and analysis assistant that shines on long documents.
Gemini
Google's assistant, strongest when your work already lives in Google Workspace.
Microsoft Copilot
AI inside Word, Excel, Outlook, and Teams, with your company's data protections.
Canva
Design for non-designers, with Magic Studio AI for text, images, video, and resizing.
Zapier
Connect thousands of apps with no-code workflows, now with AI steps, agents, and chatbots
NotebookLM
A research notebook that only answers from the sources you give it, with citations.
Cautions for nonprofit professionals
Donor names, giving histories, contact details, and anything about the people you serve (especially health, immigration, housing, or minors' information) must not go into a consumer AI tool. Never paste confidential data into a consumer AI tool unless your organization has approved that tool for that use. Check state privacy laws, your donor privacy policy, and any funder data agreements.
Models will invent a plausible statistic, a moving quote, or a composite client if you let them. In grant writing and reporting that is a credibility problem and sometimes a legal one. Verify every figure against your records, use only real stories with consent, and do not use generated images to depict clients or communities.
Some funders now ask whether AI was used in a proposal or restrict it; read each RFP. Reports to government funders carry certification requirements, and Form 990 narratives must be accurate. Keep a record of what AI drafted and who verified it.
Donors give to people, not templates, and communities recognize a savior tone or a stereotype immediately. Use AI for the draft and the structure, add the personal line yourself, keep a real person's name on everything donor-facing, and have staff or community members review external language.
Your 30-day plan
- Week 1: Write the one-page data rule (what never goes into AI tools) and share it with staff and volunteers. Set up an approved tool and use it daily for emails and summaries.
- Week 2: Build your materials library: approved boilerplate, current outcomes with dates, best past proposals with client details removed. Load it into a project or custom GPT.
- Week 3: Draft one grant section and one impact report with the prompts above; verify every number against source data and log what you had to correct.
- Week 4: Draft your thank-you letter set. Build one automation with a review step and test it with dummy records.
- End of month: Review with your team: hours saved, errors caught, what donors or funders noticed. Decide the next two workflows.
Frequently asked questions
Can nonprofits use ChatGPT for grant writing?
Is it safe to put donor information into AI tools?
Will AI replace nonprofit fundraisers?
Do funders allow AI-written proposals?
Terms used on this page
Related roles
- AI for MarketersDraft campaigns faster, mine customer feedback for messaging, and turn reporting into a ten-minute job, without drifting off-brand or into a CAN-SPAM problem.
- AI for Small Business OwnersAI answers the one-star review calmly, drafts the month of posts, turns your voice memo into an SOP, and preps the questions for your CPA. You still make the calls, sign the checks, and own what goes out under your name.
- AI for Project ManagersAI cannot run your project, but it can draft the status report, turn a messy meeting into an action list, and pressure-test your risk register in minutes. Here is how project managers use it without losing the plot.
- AI for Content WritersAI can research, outline, draft, and edit alongside you. The writers who thrive with it use it to reach the interesting part faster, keep their own voice, and are straight with clients about how they work.