easyai › ChatGPT for nonprofits
ChatGPT for nonprofits
What it's actually good at in this job
Not “transforming your organisation.” These are the specific, unglamorous jobs that eat a nonprofit week, and the ones where it earns its keep first.
- Donor thank-yous that don't read like a form letter. The single highest-value use, because the alternative is either a template everyone recognises or a note that never gets sent.
- First drafts of grant applications. Not the final text. The draft that gets you off a blank page at 9pm.
- Turning a messy programme update into something a board can read. Same facts, reorganised, half the length.
- Plain-English versions of your own jargon. Every nonprofit has internal language that quietly excludes donors. It's very good at spotting it.
- One event, six social posts. Different lengths, different platforms, from a single description.
- Reading the long document nobody has time for. A funder's guidelines, a compliance pack, a 40-page report — summarised, then questioned.
Read this before you paste anything
Nonprofits handle information that is more sensitive than most businesses do, and the usual advice glosses over it. Do not paste donor records, beneficiary names, safeguarding notes, case files, or anything you hold about a vulnerable person.
If your organisation has a data protection policy, AI tools sit under it. Worth five minutes with whoever owns that policy before this becomes a team habit rather than your habit.
Try it now: a donor thank-you that sounds like a person
This is how our lessons work — you don't watch a video, you paste this into ChatGPT and it walks you through it, one step at a time, waiting for your answer at each step. Copy everything in the box.
You are my patient tutor. I work at a nonprofit and I am NOT technical.
Teach me by having me do one thing at a time, not by explaining.
Rules for this whole conversation:
- ONE step. Then STOP and ask what happened. Wait for my answer.
- No jargon. Everyday comparisons only.
- If I say "it didn't work", ask what I saw. Do not move on.
- Under 150 words per message.
- Never ask me for a donor's name, address, or any personal details.
If I paste any, tell me which parts weren't needed.
Lesson: write a thank-you to a real donor that sounds like me.
1. Ask me about one real donation, WITHOUT any identifying details -
roughly how much, what appeal, and anything I know about why they gave.
2. Ask me to paste something I have written before, so you can hear
my voice. A newsletter, an email, anything.
3. Write the thank-you. Then STOP and ask if it sounds like our
organisation or like a charity form letter.
4. Fix the one thing I name, without starting over.
5. Show me what made the difference.
Start with step 1. Ask me now, and nothing else.
Step 3 is the whole lesson. If the draft comes back sounding like your organisation, you have just learned in ten seconds what a paragraph of explanation could not teach you.
What it will get wrong
- Impact numbers. It will produce confident, wrong figures if you let it. Every number in a grant application or annual report comes from your records, never from the model.
- Funder requirements. It does not know a specific funder's current criteria. Check the guidelines yourself.
- Your tone, at first. Out of the box it writes like a corporate brochure. That's what step 2 above is for.
- Anything legal or regulatory. Charity law, tax treatment of gifts, safeguarding duties — those go to a person who is qualified and insured.
Questions
Can nonprofits use ChatGPT for free?
Yes. Everything on this page works on the free plan. You do not need to pay anything to draft thank-you letters, summarise documents or get a first pass at a grant application.
Is it safe to use ChatGPT with donor information?
Don't paste it. Donor records, beneficiary names, case notes and safeguarding information should stay out of any AI tool. Describe the situation instead of naming the person: it almost never needs the identifying details to help you, and you get the same quality of draft either way.
What is the single best first use for a small nonprofit?
Donor thank-you letters. It is high volume, it matters emotionally, and the usual alternative is a template that donors recognise as a template. It is also low risk, because nothing factual depends on it.
Can ChatGPT write our grant applications?
It can write a first draft that gets you off a blank page, and it is genuinely good at that. It cannot supply your impact figures, and it does not know a specific funder's current criteria. Treat the draft as a starting point you then make true.
Will funders mind if we used AI to write the application?
Policies vary and some funders now ask directly, so check the guidelines. The practical answer is that a draft you have rewritten, fact-checked and put in your own voice is your work. A draft you pasted in unread usually reads like one.
Do we need a policy before staff start using it?
You need one before it becomes a team habit. The minimum is a short list of what must never be pasted in, which for most nonprofits means donor data, beneficiary data and anything covered by safeguarding. That can be one page.
Want the rest, free?
Six short lessons you paste straight into the tool. One real win in your first ten minutes.
Get the free kit