ChatGPT for Nonprofits [2026 Pricing, Prompts, and Setup]

Last month, our content strategist drafted a 2,500-word blog post in 90 minutes. Two years ago, that same kind of post took her a full day. The difference was one tool and one skill: ChatGPT and knowing how to prompt it.

I'm Silvia Li Sam, founder of Slam Media Lab. My team uses AI every single day to write content, build marketing strategies, analyze data, and manage projects. We've tested every major AI tool on the market. We know which ones are worth your time, which ones waste it, and how to set them up so your nonprofit team actually uses them.

Here's the reality: 92% of nonprofits are now using AI in some form, according to a 2026 Virtuous report that surveyed 346 organizations. But 81% are using it ad hoc with zero documentation. Only 4% have AI-specific training budgets. And 60% say they lack the expertise to even evaluate which tools to use.

This guide closes that gap. I'm going to walk you through exactly how to use ChatGPT for nonprofits in 2026, with actual prompts you can copy, pricing breakdowns, the AI tools most guides skip, and the data privacy rules your board needs to know.

‍

How Does ChatGPT Work?

ChatGPT is an AI chatbot built by OpenAI that generates text by predicting the most relevant response based on patterns learned from billions of pages of text. You type a prompt. It responds with human-like text.

The model powering ChatGPT in 2026 is GPT-5. It handles longer documents, reasons through multi-step problems, and produces more accurate output than GPT-4. It can also browse the internet, analyze uploaded spreadsheets, generate images, and build simple automations.

If you want the backstory on OpenAI's nonprofit origins and how the company evolved, we wrote a deep dive on that.

ChatGPT Pricing for Nonprofits in 2026

Pricing is the first question every nonprofit asks, and it's changed a lot since 2024. OpenAI restructured its tiers in early 2026, renaming the Team plan to Business and adding a new budget-friendly Go tier. The good news: there's now an official nonprofit discount program that didn't exist a year ago. Here's what each tier costs, what you get, and which one makes sense depending on your team size and budget:

• Free ($0/month): GPT-5 with limited messages, slower image generation, and basic deep research. Good for one staff member testing the waters.
• Go ($8/month): More capacity than Free, ad-supported. A low-cost entry point if you need more than free but aren't ready for Plus.
• Plus ($20/month): Ad-free, expanded messaging, faster image creation, full deep research and agent mode. Best for your power user, whether that's a communications director, development lead, or program manager who uses AI daily.
• Pro ($200/month): Unlimited everything, including o1 Pro mode for complex reasoning. Best for orgs doing heavy grant research, data analysis, or document processing every single day.
• Business ($25/user/month annually, $30 monthly): Unlimited GPT-5, team workspace, custom GPTs, admin controls with SSO, and the big feature: your data is never used to train OpenAI's models. Minimum two users. This is the tier most nonprofits should start with.
• Enterprise (custom pricing): Everything in Business plus extended context, SOC 2 compliance, data residency, and 24/7 support. For large nonprofits with 50+ staff or strict compliance needs.

The ChatGPT Nonprofit Discount

OpenAI launched its OpenAI for Nonprofits program with real discounts (apply through TechSoup or Goodstack):

• Business tier: 20% off (roughly $20/user/month)
• Enterprise tier: Up to 50% off (some sources cite up to 75% for larger organizations)
• Who qualifies: Registered 501(c)(3) or international equivalent. Academic, medical, religious, and government institutions are not eligible.
• How to apply: US nonprofits verify through TechSoup or Goodstack. International orgs contact OpenAI sales directly.

At roughly $20/user/month with the discount, Business gives you team workspaces, custom GPTs, admin controls, and data privacy protections. That's the tier I recommend for most nonprofits.

AI Tools for Nonprofits Beyond ChatGPT

ChatGPT gets all the headlines, but three other AI tools for nonprofits deserve your attention. All of them offer nonprofit discounts.

Claude (by Anthropic)

Claude is the AI tool my team uses most for long-form work. It handles documents up to 200,000+ words in a single conversation, writes with more nuance than ChatGPT, and excels at detailed analysis. If your nonprofit produces lengthy grant applications, policy documents, or annual reports, try Claude first.

• Free: Limited daily messages
• Pro: $20/month (5x free tier usage)
• Max: $100 or $200/month for heavy users
• Team: $25/user/month standard, $150/user/month premium
• Enterprise: Custom pricing

The nonprofit deal: Anthropic launched Claude for Nonprofits with up to 75% off Team and Enterprise plans for 501(c)(3) organizations, K-12 schools, and qualifying healthcare orgs. The plan includes connectors to Blackbaud, Candid, and Benevity, plus a free "AI Fluency for Nonprofits" training course. Verify through Goodstack.

Google Gemini

If your nonprofit uses Google Workspace (and most do), this is the easiest and cheapest AI to adopt. Google for Nonprofits gives eligible organizations Gemini free for up to 2,000 users. That includes the standalone Gemini app with Deep Research, NotebookLM, and Gemini built into Gmail, Docs, Sheets, and Meet with enterprise-grade data protections.

You don't need to change any tools. AI just shows up inside the apps your team already uses. Advanced Workspace plans are available at 70%+ discounts, starting at $3.50/user/month.

Microsoft Copilot

If your nonprofit runs on Microsoft 365, Microsoft for Nonprofits offers:

• 75% discount on Microsoft 365 Business Premium
• 15% discount on Microsoft 365 Copilot
• $2,000/year in Azure credits
• Free security risk assessments

Copilot integrates into Word, Excel, PowerPoint, Outlook, and Teams. Twenty-three percent of nonprofits already use it, making it the second most popular AI tool after ChatGPT (57%).

Which AI Tool Should Your Nonprofit Pick?

• Tightest budget: Google Gemini. Free for nonprofits with enterprise data protection included.
• Best for writing and long documents: Claude. Up to 75% off for nonprofits.
• Most versatile with custom bots: ChatGPT. 20-50% off for nonprofits.
• Already on Microsoft 365: Copilot. 15% off plus $2,000 Azure credits.
• Best free-tier data privacy: Google Gemini. Enterprise protections included at $0.

How to Actually Get Expert-Level Results From AI

Most guides tell you to "be specific with your prompts." That's like telling someone to "play better" at tennis. It's not wrong. It's just not useful enough to change anything.

I use AI tools 6-8 hours a day. I've done this for two years. Here's what I've learned: there are five levels to getting good output from ChatGPT, and most nonprofits are stuck on level one.

Level 1: Be Specific (Where Most People Stop)

Instead of "write a fundraising email," you include your org name, audience, tone, word count, and specific details.

Lazy prompt: "Write a fundraising email for our nonprofit."

What you get: Generic, could-be-anyone copy. Useless.

Specific prompt: "Write a 200-word fundraising email for our spring gala on April 12. We're the Bay Area Literacy Project. The audience is returning donors who gave $500-$2,000 last year. Tone: warm, personal, grateful. Mention that last year's gala raised $180,000 and funded 340 tutoring sessions. Ask for $1,000. Close with a personal note from our ED, Maria."

What you get: A solid first draft you can edit in 10 minutes instead of writing from scratch in 45.

That's a good start. But it's still just level one.

Level 2: Feed the AI Your Organization's Context

This is where most nonprofits never go. Before you ask ChatGPT to write anything, give it your raw materials.

Paste in your brand voice guidelines, or 3-5 examples of communications your ED loves. Tell it: "Here are five emails our executive director wrote that represent our voice. Study them. Match this tone in everything you write for us."

Upload your annual report. Tell it: "Read this. You'll reference these metrics and stories when I ask you to write donor communications."

Paste in last year's highest-performing fundraising email, your best blog post, or the grant narrative that won funding. Tell it: "This got results. Use it as a model."

When you do this, ChatGPT stops writing generic nonprofit copy and starts writing like someone who actually works at your organization. The difference is night and day, and it takes about five minutes of setup.

Level 3: Make the AI Research Before It Writes

This separates amateur prompting from a professional workflow. Before asking AI to produce anything public-facing, make it do homework first.

Here's a real example from how we work at Slam. When I need a blog post on a topic, I don't say "write me a post about nonprofit SEO." I say:

"Before you write anything, I need research. Look at what's currently ranking for [KEYWORD]. Tell me: What are the top five articles covering? What angles are they missing? What questions are people asking that nobody answers well? What data or case studies can we include that competitors don't have?"

Then I review that research. Then I tell the AI what angle to take. The writing comes last, not first.

Your nonprofit can do the same thing:

• Before a fundraising appeal: "Research what the top-performing nonprofit fundraising emails of 2025 had in common. What subject lines, structures, and CTAs drove the highest conversion?"
• Before a grant narrative: "Research [FOUNDATION NAME]. What are their stated priorities? What have they funded in the past two years? What language do they use in their own materials?"
• Before a social media campaign: "Look at what other [CAUSE AREA] nonprofits are posting on Instagram right now. What's getting engagement? What's falling flat? What could we do differently?"

The AI does 30 minutes of research in 60 seconds. Then you make the strategic decisions based on what it found.

Level 4: Make the AI Interview You First

This is the technique that changed how I work, and almost nobody talks about it.

Instead of trying to stuff everything into one perfect prompt, tell the AI to ask YOU questions before it starts writing.

Try this prompt:

"I need you to help me write [TYPE OF CONTENT] for our nonprofit. Before you write anything, ask me 10 specific questions about our organization, audience, goals, and constraints so you fully understand the situation. Ask them one at a time and wait for my answer before moving to the next."

What happens next is powerful. The AI asks things you wouldn't have thought to include: What's your audience's biggest objection? What's been tried before that didn't work? What does your board care about most? What's the one action you want people to take?

By the time you've answered 10 questions, the AI has deep context. The output it produces after that conversation is dramatically better than anything from a single prompt, no matter how detailed. It's like the difference between handing a freelancer a one-line brief vs. having a 30-minute kickoff call.

We use this approach at Slam for almost everything complex: content strategy, campaign planning, brand messaging, competitive analysis. The AI becomes a strategic thinking partner, not a text generator.

Level 5: Iterate Like a Colleague, Not a Vending Machine

The biggest mistake nonprofits make with AI: treat it as a one-shot tool. Type a prompt, get output, copy-paste, done. That leaves 80% of the value on the table.

Here's how we actually work at Slam:

1. Get the first draft. It won't be right. That's fine.
2. Give specific feedback. Not "make it better." Say: "The second paragraph is too formal. Rewrite it like you're explaining this to a friend over coffee. The CTA is buried at the bottom. Move it to right after the success story."
3. Add what's missing. "Good, but you didn't mention our new after-school program. Add a paragraph about it between sections 2 and 3, using these metrics: [PASTE DATA]."
4. Challenge it. "The opening isn't strong enough. Give me five alternative first lines that start with a specific number or a surprising fact."
5. Final polish. "Read through this entire draft. Flag anything generic, any claim without a specific example, and any sentence longer than 25 words. Then fix all of them."

This back-and-forth turns AI from a mediocre first draft machine into a genuine thinking partner. It's how we produce work at Slam that sounds like a human wrote it, because a human directed every revision.

I'm going to give you copy-paste prompts for every use case below. Customize the brackets. Build a shared document your whole team can pull from.

10 Ways to Use ChatGPT at Your Nonprofit (With Prompts You Can Copy)

1. Fundraising Emails and Donor Thank-Yous

The fastest win for most nonprofits. Donor communications eat hours every week, and ChatGPT can cut that time by 60-80% when prompted right.

Copy this prompt:

"You are a development director at [ORG NAME], a [MISSION] nonprofit in [CITY]. Write a thank-you email to a donor who just gave $[AMOUNT] to our [CAMPAIGN]. Tone: warm, specific, and personal. Mention how their gift will fund [SPECIFIC IMPACT, e.g., '12 after-school tutoring sessions']. Keep it under 150 words. Sign off from [ED NAME], our executive director."

Why it works: It gives ChatGPT a role, an audience, a tone, a specific impact detail, and a word count. Every one of those constraints makes the output better.

The numbers back this up: nonprofits using AI-assisted donation forms see an average gift of $161 vs. the $115 industry average, a 40% increase (Fundraise Up 2025).

2. Blog Posts and Content Marketing

At Slam, AI handles the first draft of every blog post. A human handles the voice, the fact-checking, and the strategy. That combination is what makes it work. Fifty-three percent of nonprofit communicators already use AI to brainstorm headlines, and 39% use it for first drafts (Nonprofit Tech for Good).

Copy this prompt:

"Write a blog post outline for [TOPIC] targeting [AUDIENCE]. The post should be about 2,000 words, written in first person from [AUTHOR NAME/TITLE]. Include 5-7 H2 sections with specific, actionable advice. Each section should include a real example or a data point. Tone: like teaching a smart colleague, not lecturing. No jargon."

Use this to draft posts for SEO, write captions for Instagram and LinkedIn, create newsletter content, repurpose long reports into shorter formats, and plan your editorial calendar.

3. Grant Writing and Research

Grant applications are one of the highest-value, highest-time-cost tasks at any nonprofit. A single proposal can take 20-40 hours between research, drafting, revision, and formatting. ChatGPT won't write the whole thing for you (and shouldn't), but it can cut the research and drafting phases in half when you use the level 3 approach: research the funder first, then write.

Copy this prompt:

"I'm applying to [FOUNDATION NAME] for a $[AMOUNT] grant to fund [PROGRAM]. Our organization is [ORG NAME], a [MISSION] nonprofit serving [POPULATION] in [REGION]. Our key outcomes from last year: [LIST 3-5 METRICS]. Draft a 500-word narrative for the 'Program Description' section. Clear, outcome-focused language. No jargon. Weave our metrics in naturally."

Important: 23% of foundations won't accept AI-generated content, and 67% are undecided (Candid.org). Always have a human review and personalize every application. Disclose AI use if a funder asks.

4. Board Meeting Prep and Summaries

Board members cite AI's top three benefits as minutes automation (82%), document summaries (75%), and director prep (63%) (OnBoard Meetings). This saves your ED hours every month.

Copy this prompt:

"Here are notes from our board meeting on [DATE]. Organize them into: (1) Key decisions made, (2) Action items with owners and deadlines, (3) Topics tabled for next meeting, (4) A 3-paragraph summary I can email to board members who missed the meeting. Professional but conversational tone."

Pro tip: Pair this with Fathom AI (free) to auto-transcribe meetings. Paste the transcript into ChatGPT and ask for action items. At Slam, we built a custom GPT called "Slam Action Item Bot" that does exactly this. Any nonprofit can build one in under an hour.

5. Volunteer Communications and Scheduling

Volunteer coordination is the task nobody budgets enough time for. Between scheduling, reminders, training materials, feedback collection, and no-show follow-ups, your volunteer coordinator can easily spend 6-10 hours a week on communications alone. ChatGPT paired with automation tools can handle that entire pipeline, and once it's set up, it runs without you touching it.

SisterLove, a reproductive justice nonprofit, paired Zapier with OpenAI to automate their content pipeline. Meeting notes fed into AI-generated blog posts, auto-formatted, then scheduled as emails and SMS. Result: 190+ hours saved in nine months.

You can build similar workflows using:

1. Zapier (15% off for nonprofits) to connect ChatGPT to your CRM, email, and scheduling tools
2. Make.com (free 12-month license for qualifying NGOs) for visual multi-step workflows
3. n8n for a self-hosted, open-source option with no per-task fees

6. Data Analysis and Reporting

ChatGPT's Advanced Data Analysis (Plus and above) processes spreadsheets, finds trends, creates charts, and writes narrative summaries. Upload your program outcomes CSV and ask it to find the story.

Copy this prompt:

"I'm uploading our program data. Analyze it and tell me: (1) Our three strongest outcomes with specific numbers, (2) Any trends I should flag for the board, (3) Year-over-year changes worth highlighting, and (4) A 200-word narrative summary for our annual report."

THINK South Africa used this approach to train public health professionals on cleaning and analyzing health data. Same method works for any nonprofit sitting on data they haven't had bandwidth to dig into.

7. Translating Content for Multilingual Communities

ChatGPT supports 50+ languages with better context and cultural nuance than basic translation tools. For nonprofits serving diverse communities, this matters.

Copy this prompt:

"Translate the following [DOCUMENT TYPE] from English to [LANGUAGE]. Maintain a [FORMAL/CONVERSATIONAL] tone for [AUDIENCE]. Flag any idioms or cultural references that don't translate directly and suggest alternatives. Here's the content: [PASTE TEXT]"

For full nonprofit website translation, pair ChatGPT with Weglot, which integrates with Webflow and WordPress for ongoing automated translation.

8. Building Custom GPTs for Your Organization

If you only do one thing from this entire guide, do this. Custom GPTs let you create specialized AI assistants trained on YOUR organization's documents, voice, and processes. Instead of re-explaining your mission and programs every time you start a new chat, you build a GPT once and your whole team uses it forever. Three examples you can build in under an hour:

• Donor FAQ Bot: Upload your annual report, impact data, and program descriptions. Staff answer donor questions instantly without digging through files.
• Onboarding Assistant: Upload your employee handbook and policies. New hires ask questions instead of waiting for HR.
• Grant Research Bot: Upload your mission statement, past successful applications, and target foundations. It matches you to relevant opportunities and drafts initial outreach.

How to build one: In ChatGPT (Business tier), click "Explore GPTs" then "Create." Name it, write 2-3 sentences of instructions, upload your documents, test it. That's it. Thirty to sixty minutes, and your team has a custom AI assistant.

9. Crisis Response and Communications

The Trevor Project built a Crisis Contact Simulator that trains crisis counselors with lifelike AI conversations. TIME named it a Best Invention of 2021.

For most nonprofits, the practical use is faster crisis comms. Feed ChatGPT your crisis playbook, then ask for:

1. An initial public statement within minutes
2. Internal staff updates with talking points
3. Donor communications explaining your response
4. Social media posts matched to the right tone
5. A media FAQ

Copy this prompt:

"Our nonprofit [ORG NAME] is responding to [BRIEF SITUATION]. Our values are [2-3 VALUES]. Draft a 150-word public statement: transparent, empathetic, action-oriented. Include what we know, what we're doing, and how stakeholders can get updates. No speculation."

10. SEO and Getting Found by AI Search

Something most ChatGPT for nonprofits guides skip: AI is changing how people discover organizations online. ChatGPT, Gemini, and Perplexity are answering questions that used to send people to Google. If your content isn't structured for AI citation, you're losing a growing discovery channel.

This is called Generative Engine Optimization (GEO), and we specialize in it at Slam alongside traditional nonprofit SEO. It means writing with clear statements, specific metrics, and structured content so AI models can parse and recommend your organization.

ChatGPT can help you research keywords, draft SEO-optimized pages, find content gaps, generate meta descriptions, and build FAQ sections that show up in AI answers.

How to Set Up ChatGPT at Your Nonprofit (Step by Step)

I've watched nonprofits fail at AI adoption because they tried to change everything at once. A communications director signs up for ChatGPT, sends a Slack message saying "hey everyone, we have AI now," and three months later nobody uses it. The organizations that succeed follow a deliberate process. It takes about a month, and it works because it builds proof before it asks for buy-in.

1. Pick one team. Communications or development usually has the clearest repetitive pain points.
2. Choose your tier. Budget tight? Start with Google Gemini (free). Can invest $20/user/month? ChatGPT Business with the nonprofit discount.
3. Turn off data training. Free/Plus tiers: Settings > Data Controls > toggle off "Improve the model for everyone." Business/Enterprise: off by default.
4. Build your first custom GPT. Start with a donor FAQ bot. Upload your annual report. Test it with 10 real donor questions.
5. Create a prompt library. Shared Google Doc or Notion database. Every time someone writes a prompt that works, add it. Organize by department.
6. Train hands-on. Sit with each person. Walk through their top 3 repetitive tasks. Build the prompts together. Show the before and after. This is not a slide deck moment. It's a "sit next to me and watch" moment.
7. Measure month one. Track hours saved per person per week. That number is what convinces your ED and board to expand.

Data Privacy: Protecting Donor and Client Information

Seventy percent of nonprofits are concerned about data privacy with AI (JRF Report). They should be. Here are the three rules that matter:

Rule 1: Never paste donor PII into a free-tier AI tool. Free tiers may use your inputs to train future models. That means donor names, addresses, emails, and giving histories could end up in training data. Only use Business or Enterprise tiers for sensitive information.

Rule 2: Create an AI governance policy. Seventy-six percent of nonprofits don't have one (TechSoup 2025). Your policy should cover which tools are approved, what data can and cannot be entered, who reviews outputs before they go external, and how you disclose AI use.

Rule 3: Know your compliance requirements. Serving youth? COPPA. Health data? HIPAA. EU clients? GDPR. Consult legal counsel before entering client or beneficiary data into any AI system.

Which tiers protect your data:

• ChatGPT: Business and Enterprise. Data not used for training. ~$20/user/month with nonprofit discount.
• Claude: Team and Enterprise. Data not used for training. As low as ~$6/user/month with 75% nonprofit discount.
• Gemini: Free Workspace for Nonprofits includes enterprise-grade protections. Best option if budget is your primary concern.
• Copilot: M365 Business and Enterprise controls. 15% nonprofit discount.

What We've Learned Using AI Every Day at Slam

I want to be straightforward: we started using AI tools for ourselves before we ever offered it as a service. We tested everything. We broke things. We figured out what actually works.

Here's what two years of daily AI use across a 10-person agency taught us:

• AI saves us 15-20 hours per week across content writing, SEO research, client communications, meeting summaries, and project management.
• The prompt library is the biggest unlock. We spent months building ours. Most teams skip this step and then wonder why AI outputs sound generic.
• No AI tool replaces human judgment. Every output gets a human review. Every claim gets fact-checked. Every client deliverable gets a final human edit. AI handles the first draft and the research. Humans handle strategy, voice, and quality.
• Training is the hard part, not technology. Getting people comfortable, building habits, showing them the before and after. That takes sitting next to someone and doing it together, not sending a link to a webinar.

We're now offering AI implementation for nonprofits because we've done it ourselves and measured the results. If your team wants help setting up ChatGPT, Claude, Gemini, or automation workflows, schedule a free consultation and we'll walk through what's realistic for your organization and budget.

‍

Frequently Asked Questions About ChatGPT for Nonprofits

Is ChatGPT Free for Nonprofits?

The free tier is available to everyone. OpenAI also offers 20% off Business and up to 50% off Enterprise through the OpenAI for Nonprofits program. US nonprofits verify through TechSoup or Goodstack.

Which AI Tool Is Best for Nonprofits on a Tight Budget?

Google Gemini through Google for Nonprofits. Free for up to 2,000 users with enterprise data protections included. It works inside the Google Workspace apps your team already uses.

Can Nonprofits Use ChatGPT for Grant Writing?

Yes. It can draft narratives, generate logic models, and edit for word count. But 23% of foundations won't accept AI-generated content, and 67% are undecided (Candid.org). Human review on every application. Disclose if asked.

Is It Safe to Use Donor Data With ChatGPT?

Only on paid tiers where data isn't used for training. ChatGPT Business/Enterprise, Claude Team/Enterprise, and Google Gemini for Nonprofits all qualify. Never enter donor names, addresses, or giving amounts into a free-tier tool.

How Do I Get My Team to Actually Start Using AI?

One team. One pain point. One tool. Train hands-on. Measure hours saved in the first month. Once one team shows results, the rest will want in. If you need help, get in touch with us about nonprofit AI implementation.

How Is AI Changing How People Find Nonprofits?

AI tools like ChatGPT, Gemini, and Perplexity answer questions that used to drive Google search traffic. Nonprofits that structure content for AI citation (clear statements, specific data, structured formats) get recommended by these tools. This is called GEO (Generative Engine Optimization), and it's becoming as important as traditional nonprofit SEO. Slam is one of the few agencies that offers both.