Community IT Innovators Nonprofit Technology Topics

Nonprofit AI: Introduction with Carolyn Woodard

• Community IT Innovators • Season 7 • Episode 2

Share episode

Community IT is starting a new series today: Nonprofit AI. Midweek we will share 10-15 minutes of updates about AI and nonprofits - including current news stories, tips, definitions, use cases, frameworks, and resources.

Whether you are an AI novice wondering how to catch up quickly, or an AI early adopter always looking for a new tool you can use at your nonprofit, we hope you will join us every Tuesday for another quick update on what is going on in the world of Nonprofits and AI. 

Takeaways and resources from Ep 1:

1. How AI Works: The Fast-Paced Library

Think of a Large Language Model (LLM) like a super-fast librarian who has read almost everything ever written. When you ask a question, the AI doesn't "look up" a file; it predicts the next word in a sequence. It processes your request into small "packets" of data (tokens) that are sent to massive datacenters. There, billions of mathematical calculations happen in milliseconds to return a response that sounds human.

2. Embedded AI vs. Prompting AI

You are already using AI, even if you haven't opened a chatbot.

  • Embedded AI: This is "hidden" technology inside tools you use daily, like Google Search algorithms, GPS route optimization, or even your email’s spam filter.
  • Prompting AI: This is "Generative AI" like ChatGPT or Gemini, where you actively start a conversation (a "prompt") to create something new, like a draft email or a report summary.

3. Use Enterprise Logins

You should use the Enterprise versions of tools like Microsoft Copilot or Google Gemini (logging in with your work account) rather than the free, public versions. This should keep your data "walled off." This ensures your donor information or internal notes aren't used to train the public model or seen by anyone outside your organization.

4. Policies are Your Starting Point

While many nonprofits are still catching up on formal IT governance or employee handbooks, AI represents a unique moment to start documenting your "rules of the road." You don't need a 50-page document, but you do need clear guidelines for your team on what data can be shared with AI, who is responsible for fact-checking AI outputs, and how your organization discloses AI use. 

5. It is Okay Not to Know Everything

Your role as a leader is to focus on strategy and ethics, not the underlying code. It is perfectly professional to say, "I'm still evaluating how this tool fits our mission," or "I need more information on the privacy implications before we proceed."

Resources:

What is Generative AI? – IBM

AI for Nonprofits: What You Need to Know – In the Microsoft/TechSoup Digital Skills Center

AI Is Already in Your Nonprofit – Community IT Innovators

AI Suitability Kit for Nonprofits - NetHope

Artificial Intelligence (AI) for Nonprofits – The Nonprofit Alliance

_______________________________
Start a conversation :)

Thanks for listening.


Carolyn Woodard:

Hello and welcome to Community IT Innovators Technology Topics Podcast. My name is Carolyn Woodard. I'm your host, and today we are starting a new journey in 2026 where I will be doing a midweek update on nonprofits and AI. Artificial intelligence is on everyone's mind these days, and it really is impacting nonprofit IT. If you don't think you know everything about AI, join the club. Even AI experts don't know everything that's going on with every tool all the time and all the implications. So this year we're going to try this experiment of a short midweek podcast, just a little news update on something about AI that is in nonprofit news, maybe a resource or two to point you to, a helpful tip, some other ideas. Maybe we'll have some guests, but in general, it'll just be me doing a quick update. So welcome. If you are new to AI, if you are accomplished and happy and always learning new AI tools, please welcome everyone. We're going to learn a lot more about this together, especially how it interacts with nonprofit IT.

Carolyn Woodard:

So I wanted to start out our first day with the AI podcast to just maybe go back and go over definitions and kind of set some levels. So if you are not at all sure what AI is, even though it's all around all the time, don't worry, you are not alone. I was talking with someone over the holidays and she just asked me, you know, what is AI? I keep hearing about it. So it's not true that you're the only one who hasn't been paying attention. So I think also one of the things about artificial intelligence AI is that we have been using it for a long time, just in different guises. And so, you know, when you use GPS in your car, uh when you are doing Google search and you're getting suggestions, the algorithm is, you know, using these types of artificial intelligence uh analysis to return the things that you're most likely to be looking for. Um so all of that is AI. Uh

Carolyn Woodard:

I talked to someone recently who said it's uh convenient to think about it in buckets of that you have the embedded AI, and this is something that we've had for a while, like in GPS, like in the algorithms. And that's gonna continue, and you're going to have embedded AI in a lot of the tools that you already use. So next time you do an update for Excel, you're gonna see like, do you want to use AI for this? Um it'll be in there. Um and uh I think one of the things that really changed when ChatGPT came out and it kind of hit the big time over the past year is prompting AI. So AI where you go to Copilot or Gemini or Cloud or other AI tools, ChatGPT, although hopefully you aren't using that out in the public.

Carolyn Woodard:

If you are at a nonprofit and you are using Microsoft or Google Workspace, you probably have access to quote unquote internal AI tools like Copilot and Gemini that you can use from within your single sign-on. So it's more protected and a little bit more private. Um, of course, check your AI policy for that, or maybe check with your IT team if you're not sure how to access those tools, or if you're still going out to public Gemini or public Chat GPT, maybe talk with your IT people about it or check your policy.

Carolyn Woodard:

That prompting AI is uh those tools like a chatbot, like um uh chat GPT, where you're gonna ask it something and it's gonna go out to its database, the data center of uh other times people have asked a similar type of question and what the answers were. And you can imagine that's a lot of data. And the thing that really changed with Chat GPT, and the reason they're building all of these databases is the speed at which uh you can just put in your question prompt. It can sort through on its end what do those packets of words, when people have used those packets in those orders before, what did that sentence mean? So it crunches that like in milliseconds basically. And then it's gonna look through when that sentence has been in put in place before, what were the answers, and it's gonna go through again and find the packets. Like if I use this packet of these words together, what's the most likely next packet of words that I would use if I were an English speaker?

Carolyn Woodard:

And this goes to these large language models, all of the um content we've been putting on the web, on the internet for a decade at this point. And um, so the artificial intelligence is it's doing something very similar to what we do in our brains, um, but it's doing it in this kind of mechanical way of saying it's just going by the probabilities. So it doesn't know that this is a sentence, that it has meaning. It knows that in general, these types of words packets come after these other types of word packets, and that makes a sentence that is understandable. And of course, it's always learning. So it's getting better and better at returning something that is a sentence that makes sense in that context. Although every once in a while you'll still catch it doing something that you're like, oh, that isn't what that means at all.

Carolyn Woodard:

Which is one of the other big uh fundamental things that we are advising all of our nonprofit uh clients to do, which is make sure two things. One, that you have a human being looking at any output, um, especially any output that is critical to you. So something that is public, uh something that would have a damaging effect if it's not correct and accurate, or if it doesn't make sense, or if it's insulting. So you should always have a human being as the final editor. And that's not just on content that you're putting on your website that was maybe drafted by AI tools, but you know, Excel spreadsheets that you're using.

Carolyn Woodard:

If you ask AI to help you with it, make sure that you're doing something that you know what the answer is or should be, and you're checking it. Um, like any assistant, you're gonna give it some work to do and then you're gonna check its work. Um, so that's one thing to think about when you're using uh AI is make sure you have a human uh that's the editor.

Carolyn Woodard:

And then, of course, you want to make sure that you're using AI in compliance with your AI policies. This may be something that your nonprofit is not quite got all of your policies formally in place. Um, they may not be public policies, they may be kind of floating around in the background. Um, if you don't have an AI policy in your acceptable use policy, you know, that big policy document usually you read when you're first hired, and then you don't usually look at it very much anymore after that. But um it's a good thing to start that conversation. Ask what is the AI policy? Just keep asking. Um, people may say, oh, that's a good idea, we should have one. People may say, oh, it's on page 23. Didn't you read your employee handbook? Um, so just make sure that you're uh asking about the policy and that what you're using AI to do at your nonprofit is complying with that policy.

Carolyn Woodard:

Um so to get back to what AI is, um that's basically it. It's going out to it, it's it's using its database, its large language model, to predict, sometimes they call it predictive text, um what you are asking it to do. Um, and in a future podcast, uh AI podcast, we'll talk a little bit more about prompting and how you can write better prompts to get more accuracy.

Carolyn Woodard:

And you can prompt in ways that the AI tool will return information about what it's not sure about, which can be really useful to you when you are the human editor giving a last look over what the AI has drafted for you. Um, so we'll cover that in future. But for this episode, I wanted to just do a quick level setting. I'm gonna include some resources in the show notes. Um, this will be on our website as well all year. We're gonna be sharing these resources with you with links and um checked for accuracy. And uh hope this will be helpful to you.

Carolyn Woodard:

As always, if you have questions or comments, um you can always respond to the podcast. I see uh comments uh where you're listening. Um I would love to interact with you on LinkedIn. You can ask questions there. We're on Reddit at nonprofit IT Management. Um, so you can ask questions there as well. I will be posting all of these resources there as well.

Carolyn Woodard:

We know that AI is, as someone said, it's the water that we're swimming in now. Um you can hold your breath if you want to. And a lot of nonprofits, of course, are seeing a lot of the ethical issues around AI, the uh transformation of neighborhoods and communities, the environmental impact, the concentration of power and the power imbalance of these four companies that pretty much are running all AI.

Carolyn Woodard:

I did hear at a conference this great quotation. Someone said, Whenever you think anything about AI, remember that there is a billion-dollar advertising campaign aimed at you, the consumer, to convince you that AI is something that you should use and that is great. So, just like any other advertising campaign, like take it with a grain of salt and make your own decisions as much as you can, knowing that all of the advertising is all around you, as well as AI being all around you. Um, so I hope that was helpful.

Carolyn Woodard:

This is our midweek AI podcast as part of the Technology Topics podcast from Community IT. And I'm Carolyn Woodard, your host, and I look forward to continuing to talk about AI all year.