After two years of experimenting with AI tools in my nonprofit work, I’ve moved from exploration to integration. Here’s my journey and what I’ve learned about making AI useful in the nonprofit context.
My AI Toolkit: What’s Working
I’ve settled into a rhythm with several AI tools consistently delivering value. Grammarly‘s premium account has become my trusted writing partner, while Claude and ChaptGPT are my analytical sidekicks. Each platform offers unique strengths – I maintain dedicated project spaces in both, paying $20 monthly. It’s an investment that pays for itself in terms of time saved and quality improved. I create spot illustrations for blog posts and newsletters using DALL-E. I prefer Perplexity.ai, an AI-fuelled search engine that provides citations for results, to Gemini/Google Search or Microsoft CoPilot.
I maintain Projects within Claude for ongoing work, including editing personal essays, analyzing reports and data for client projects, and creating guidelines for more complex analyses, such as analyzing consumer data or building a licensing model. I also make custom GPTs and use a similar Projects tool within ChatGPT, which offers some functions that Claude does not, including Code Interpreter and DALL-E for spot illustrations.
My newest tool is Google NotebookLM, which I use to organize, summarize, and share research reports for a group project. It’s way more fully featured than the older Google Keep, but I am using it to keep track of published data similarly.
What’s particularly exciting is how these tools transform traditionally time-consuming nonprofit tasks. Recently, I used my custom AI setup to analyze three years of an organization’s annual reports. Instead of spending hours extracting key metrics, I had AI highlight the critical changes and trends across the years–and provide documentation and citations for all of the work. It’s not only about saving time—it’s about asking better questions and finding new patterns.
The Reality Check: What’s Still Challenging
Let’s be honest about what’s not working perfectly. My attempts to use AI for CSV file analysis have been mixed. While I’ve had some victories with budget analysis and user data aggregation, the persistent inconsistency in handling complex datasets remains a significant hurdle for reliable implementation.
Meeting transcription tools like Otter.ai and Fireflies.ai remain in my “complicated relationship” category. They promise so much but deliver inconsistently enough that I keep toggling between using and abandoning them. This experience has taught me an essential lesson about AI adoption: sometimes, waiting for tools to mature is better than forcing the adoption of unstable solutions.
Looking Forward: My 2025 AI Exploration List
The landscape keeps evolving, and I’m particularly intrigued by several tools that could be good onprofit resources:
- Loom: Several people have told me they love using this tool to make video-like instructions for tasks (like building a custom GPT).
- GlideApp: A colleague shared how they use GlideApp to create an interface over Google Sheets, and I want to try it.
- SORA: OpenAI’s new video generation tool holds fascinating potential for nonprofit storytelling and educational content – though I’m approaching it excitedly and considering ethical implications.
- Notion: Colleagues and friends are using Notion to build web resources, and I want to delve deeper into it to understand what use cases it can best support.
What This Means for Nonprofit Leaders
For Individual Nonprofit Professionals:
Start with concrete, everyday challenges. The most successful AI implementations begin with routine tasks that take time but don’t require complex decision-making. These include report summarization, outline development, and data pattern identification. Build confidence with these before tackling more complex applications.
For Those Leading Nonprofit Organizations:
Rather than rushing to implement AI tools across your organization, identify specific departments or functions where AI could add immediate value. Create space for experimentation but maintain clear guidelines about data privacy and ethical considerations. Consider starting with an AI Acceptable Use Policy; this is a great way to have constructive conversations and build guardrails.
Moving Forward Together
Successful AI adoption in nonprofits isn’t about using every available tool. It’s about selecting and implementing tools that align with your organization’s capacity, needs, and values. Start small, document what you have learned, and build from there.
As we move through 2025, I’d love to hear how you approach AI in your organization. What tools are making a difference? What challenges are you facing?
After two years of experimenting with AI tools in my nonprofit work, I’ve moved from exploration to integration. Here’s my journey and what I’ve learned about making AI useful in the nonprofit context.
My AI Toolkit: What’s Working
I’ve settled into a rhythm with several AI tools consistently delivering value. Grammarly‘s premium account has become my trusted writing partner, while Claude and ChaptGPT are my analytical sidekicks. Each platform offers unique strengths – I maintain dedicated project spaces in both, paying $20 monthly. It’s an investment that pays for itself in terms of time saved and quality improved. I create spot illustrations for blog posts and newsletters using DALL-E. I prefer Perplexity.ai, an AI-fuelled search engine that provides citations for results, to Gemini/Google Search or Microsoft CoPilot.
I maintain Projects within Claude for ongoing work, including editing personal essays, analyzing reports and data for client projects, and creating guidelines for more complex analyses, such as analyzing consumer data or building a licensing model. I also make custom GPTs and use a similar Projects tool within ChatGPT, which offers some functions that Claude does not, including Code Interpreter and DALL-E for spot illustrations.
My newest tool is Google NotebookLM, which I use to organize, summarize, and share research reports for a group project. It’s way more fully featured than the older Google Keep, but I am using it to keep track of published data similarly.
What’s particularly exciting is how these tools transform traditionally time-consuming nonprofit tasks. Recently, I used my custom AI setup to analyze three years of an organization’s annual reports. Instead of spending hours extracting key metrics, I had AI highlight the critical changes and trends across the years–and provide documentation and citations for all of the work. It’s not only about saving time—it’s about asking better questions and finding new patterns.
The Reality Check: What’s Still Challenging
Let’s be honest about what’s not working perfectly. My attempts to use AI for CSV file analysis have been mixed. While I’ve had some victories with budget analysis and user data aggregation, the persistent inconsistency in handling complex datasets remains a significant hurdle for reliable implementation.
Meeting transcription tools like Otter.ai and Fireflies.ai remain in my “complicated relationship” category. They promise so much but deliver inconsistently enough that I keep toggling between using and abandoning them. This experience has taught me an essential lesson about AI adoption: sometimes, waiting for tools to mature is better than forcing the adoption of unstable solutions.
Looking Forward: My 2025 AI Exploration List
The landscape keeps evolving, and I’m particularly intrigued by several tools that could be good onprofit resources:
- Loom: Several people have told me they love using this tool to make video-like instructions for tasks (like building a custom GPT).
- GlideApp: A colleague shared how they use GlideApp to create an interface over Google Sheets, and I want to try it.
- SORA: OpenAI’s new video generation tool holds fascinating potential for nonprofit storytelling and educational content – though I’m approaching it excitedly and considering ethical implications.
- Notion: Colleagues and friends are using Notion to build web resources, and I want to delve deeper into it to understand what use cases it can best support.
What This Means for Nonprofit Leaders
For Individual Nonprofit Professionals:
Start with concrete, everyday challenges. The most successful AI implementations begin with routine tasks that take time but don’t require complex decision-making. These include report summarization, outline development, and data pattern identification. Build confidence with these before tackling more complex applications.
For Those Leading Nonprofit Organizations:
Rather than rushing to implement AI tools across your organization, identify specific departments or functions where AI could add immediate value. Create space for experimentation but maintain clear guidelines about data privacy and ethical considerations. Consider starting with an AI Acceptable Use Policy; this is a great way to have constructive conversations and build guardrails.
Moving Forward Together
Successful AI adoption in nonprofits isn’t about using every available tool. It’s about selecting and implementing tools that align with your organization’s capacity, needs, and values. Start small, document what you have learned, and build from there.
As we move through 2025, I’d love to hear how you approach AI in your organization. What tools are making a difference? What challenges are you facing?