Community IT Innovators Nonprofit Technology Topics

AI Maturity Model for Nonprofits pt 2

• Community IT Innovators • Season 7 • Episode 54

Share episode

Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.

0:00 | 28:49

In Part 2 of the AI Maturity Model for Nonprofits webinar, Carolyn Woodard and Mimi Yeh, Engagement Director at PTKO, and George Danilovics, VP of Technology at AHIP, pick up where Part 1 left off, moving from the framework itself to the harder question of what it takes to make AI adoption become quiet, boring, and everywhere.

Mimi and George walk through what it looks like when AI stops being an initiative and becomes infrastructure, pointing to the moment when people stop announcing they used it, the way no one says "I used Outlook to send this email." The session closes with a candid Q&A on tool volatility, vendor risk, and how to address genuine ethical concerns about AI when your colleagues or community have serious reservations.

Haven't listened to Part 1 yet? Find it in your podcast feed.

This episode covers:

  • Why you should map each team separately against the five maturity stages rather than assigning your organization a single score, and how that map becomes a diagnostic you can actually act on.
  • Why AI competency training deserves the same recurring, role-based architecture as security awareness training, not a one-time brown bag.
  • Action items you can start this week and this quarter to move one team forward one stage at a time.
  • How to think about vendor risk and tool volatility when planning for long-term AI integration, and why documenting your strategy matters more than committing to any one tool, and lets you move between vendors more easily down the road.
  • Practical guidance on addressing ethical and social concerns about AI, including when those conversations belong at the board level.

Resources Mentioned:

_______________________________
Start a conversation :)

Thanks for listening. 


Community IT Intro

Thank you for joining this Community IT podcast part two. You can find part one in your podcast feed if you subscribe wherever you listen to podcasts.

Carolyn Woodard

Welcome everyone to the Community IT Innovators webinar. This is the AI Maturity Model for Nonprofits how AI Becomes Quiet, Boring, and everywhere.

Mimi Yeh

I'm Mimi Yeh and I'm an engagement director at PTKO.

George Danilovics

Hello, everyone. My name is George Danilovics. Thank you all for joining this afternoon. I am the VP of technology at AHIP.

Mimi Yeh

So let's do a short reframe in order to remove some anxiety. Again, not only is the goal not for everybody to reach that pioneering stage. I think it's also important to consider that when people look at models like this, the first question that comes to mind is where is my organization? Where do I sit on this maturity model? And that might not be the right or the best question to ask.

Mimi Yeh

Something that we really advise leaders to think about when it comes to AI is to try not to give your organization a single maturity score because your organization is not going to use AI in the same way in all of your different departments and teams. Your finance team might be way ahead of your HR team. Marketing might be experimenting, programs might not have started yet. And that is totally normal.

Mimi Yeh

This is an uneven picture of the organization's progress, but it is not a reflection of failure. And it's not a reflection of poor planning. That unevenness is actually quite useful. When you map each team separately, you get a diagnostic rather than just a grade. And that diagnostic helps you see exactly where your internal success stories are for AI and which teams might be able to mentor others in their AI journey based off of their experiences and based off of the lessons that they've learned.

Mimi Yeh

Your fastest moving teams are the best trainers and advocates. They are the ones who know AI and they know AI in the context of what your organization does and the constraints and opportunities facing your organization.

Mimi Yeh

So if you were to do a self-evaluation exercise against this maturity model, we would say resist the urge to produce one number or one stage for the whole organization and instead produce a map, because that map is something that you can act on with real next steps. I think it is time for another poll.

Carolyn Woodard

It's time for our next poll. So talking about maps and where you are. So this one is actually multiple choice. So you can go ahead and choose as many as apply to you.

Carolyn Woodard

So the question is which teams at your organization are using AI on a regular basis? And I'm gonna be so interested to see this because I have in mind what I think it's gonna be, and I want to see what happens.

Carolyn Woodard

So the teams that you uh can identify as using AI on a regular basis are membership, marketing, fundraising, development, um, finance, IT, customer service, programming, something else. So if you have another team that is using AI, please put that in the chat. And uh not applicable again.

Carolyn Woodard

And I just, you know, given that we were just uh we didn't talk about it, but in chat, there were some questions about if you're a health-oriented nonprofit, and so you have like additional compliance rules. Um, and then how can you evaluate if the AI is going to still be in compliance with the rules that govern you or with your own policy? And uh, it's a thorny question. So we were talking a little bit about it in chat, but I think that kind of reflects on this as well.

Carolyn Woodard

So you, as you were saying, Mimi, you might have a team that is super cautious because they have more risks, and you may have a team that's really um an early adopter because they're what they're doing and their risks might be uh lower. So

Carolyn Woodard

George, can you tell us the results?

George Danilovics

The results. Uh so not actually, there is one on here that's surprising.

George Danilovics

So not surprising is uh the leaders uh of this poll of using AI with uh we've had IT at 45%, marketing right behind at 35. Um, finance was the number three. Um, so I'm curious about finance. Um

George Danilovics

Also not surprising to see fundraising and programming coming in at a tie. I am surprised to see membership pretty low, or maybe that's just the nature of not of a lot of membership organizations on this call. Um,

George Danilovics

And then also customer service came in pretty low. Um, only one person out of the 40 said customer service. And I think that's probably due to you know, customer service one being customer facing. And a lot of organizations are still hesitant about letting AI speak on behalf of them. Um, so I'm not too surprised seeing customer service still um as low as it is in the nonprofit sector.

Carolyn Woodard

This is so interesting. Thank you, everyone who filled out the poll. Just thank you, everyone, for doing that, and people who are putting in the chat what they're using AI for in their team. So thanks for sharing that with us.

George Danilovics

So, our slide, you know, our poll showed a pretty good spread. And as enough teams reach systemic, that infrastructure becomes shared. You've got prompt libraries that people are referencing, and and new hires, when they join the organization, they learn that AI is part of the organization by default. Um, they they get AI training and AI expectations right after they get their employee handbook. Um,

George Danilovics

But my favorite tell of when you get that critical mask is linguistic. And you've heard Mimi and I talk about this previously. Listen to how people talk. In the early stages, everybody announces, I used AI to draft this, I used Claude to look into this research.

George Danilovics

People stop talking about AI. Um, when's the last time you heard somebody say, I used Outlook to send this email? Um, I actually don't think I've ever said that in my career. Um, so when AI stops being remarkable in that literal sense of people not remarking about it, that's when you've matured from that initiative to infrastructure. That's the quiet boring everywhere that we've been talking about.

Mimi Yeh

And I think along those lines, you know, if AI is a required business skill in your organization, and I think the polls have proven that it if it is not, it's quickly going to be. Then developing that skill is a leadership responsibility that sits across the organization.

Mimi Yeh

We really can't hold our staff accountable for building skills when we didn't create a deliberate plan to build it with them and in them. So we want training to be more than a brown bag, a lunch and learn, a webinar like this, which is still great, but um, might need to be a more robust training program. Things like role-based training that actually tie to workflows and goals and challenges that organizations face today.

Mimi Yeh

So we would want to think about use cases and scenarios so that the finance team learns about applying AI for finance work. It's also around making sure that there's protected time on people's schedules to practice using AI. It's about maintaining a prompt library, which you may recognize previously was a marker of the systematic stage.

Mimi Yeh

And a big part of it, and I think this was one of the questions or comments that was in here, is about regularly revisiting and refreshing, because as the tools change, and they do change constantly, that requires the skill set to change, and it requires our perspective and mindset about AI to change.

George Danilovics

And, you know, here's the comparison that I'll draw for folks is security awareness training. And I hope it's something that you're all already doing, and nobody treats it as optional, um, and nobody treats it as a one and done. It's a structured training, it's a recurring training. It might even be roles-based in your organization. And we accepted that a long time ago, that that's the way it has to be.

George Danilovics

AI competency deserves that same attention and architecture.

George Danilovics

So, what happens next? Systemic, as we're moving through that systemic phase. Pilots have end dates, operations don't. A pilot can succeed and still strand you because nobody planned for the day after the pilot. Ownership moves from those early champions to managers. Your champion got you here, and you need to acknowledge and thank them.

George Danilovics

But one enthusiastic person can't scale. The durable version of this is managers embedding AI into how their teams work. And then the last is the scorecard changes. Success stops being measured in the number of tools tried, a number of pilots that we started, and we begin to look at how we're measuring productivity.

George Danilovics

If you take one thing from this slide, think of this. For each AI effort in your organization, can you name the manager who owns it after that pilot ends? If your answer is that person who was very excited that ran the pilot, you're still in that early experimental stage.

Mimi Yeh

Yeah, that's true.

Mimi Yeh

So we want to help you continue with your AI maturity journey. And we want to make this actionable before we go to your questions. And you see that this slide is very small, it's very simple, and it's that way deliberately. That's when the programs will stall. Narrower, achievable goals can be more easily accomplished, and they can also create the momentum to move forward.

Mimi Yeh

Organizations don't change overnight. The ones that make the most progress are usually taking those incremental, deliberate steps.

Mimi Yeh

So you can see this week, we recommend that you do just two things. Map your teams against those five stages that we showed earlier. And remember, we're talking about your team's plural, not your one organization. So you want the map. It won't take you a long period of time. And it'll probably give you more insight and more perspective on what to do next than most assessments would out there.

Mimi Yeh

And then sometime this quarter, two more things to do. Pick one team and decide the steps that you would want to take to move that team from its current stage to the next stage. And that's it. Just one team at a time, one stage over 90 days. And maybe while you're at it, a little stretch goal for people, maybe replace one brown bag with one role-based training session that is tied to a team's actual work.

Mimi Yeh

When you do that slowly and incrementally, you can repeat that play a few more times, and then you'll be a different organization within a year. So this is really about changing, learning, and building gradually in order to create capability within your organization. And it's a capability that will compound over time.

Carolyn Woodard

I love that framing of it for sure, because I think to a lot of people, AI is like this giant tidal wave that's washing over all of us. So finding these small ways to move that pilot or that team into more, you know, regular, quiet, boring, and everywhere, I think that's a great way to think about it. Um,

Carolyn Woodard

So here's our contact information, but I'm gonna move on to this slide, which has the contact information for uh Mimi and George. So you can follow them on LinkedIn and download the AI journey guide and supplemental resources at this QR code. Um and then we did have we had some questions from registration, but

Carolyn Woodard

I want to go back to a really interesting question I found in chat that we didn't ask at the time. But um, someone said in the chat that it's hard to say what it would look like if AI was as unremarkable as email, because right now the AI tools don't seem as stable as email. So the vendors could go out of business, the tools could change dramatically. I feel like every time I open Claude, there's another update. So it's hard to consider integrating AI tools when the tool itself might not exist in a year because we're in this bubble and there's lots of market consolidation going on.

Carolyn Woodard

So, how do you plan for that sort of thing when you're thinking about this bigger picture of really integrating AI into workflows, the way that Outlook is or the way that spreadsheets are?

George Danilovics

I'll take uh the first step at this and then I'll let Mimi chime in. I think when you're looking at tools, you know, you're right. Everybody is talking about AI, and there's a new vendor and a new tool every day. Um, so like any technology initiative, you need to do a little due diligence. Um, you need to look at the platform, look at the vendor, and see if that's somebody that you want to be doing business with. Um, and you you've got to take that risk and start to engage with that vendor and to see how their product fits into your organization and the work that you do. Um,

George Danilovics

You know any vendor you work with could go out of business. It's a risk that we deal with on a regular basis. But thinking about what it looks like, um, it's actually something that you won't know it or you won't see it until after it happens. Um, and

George Danilovics

And here's uh here's a personal example, and I'm even going to draw on uh a comment here that Greg had in the chat. Is you know, my AI journey probably began about two years ago, and I used um uh ChatGPT to create pictures of pandas because I couldn't think of what do I use AI for. So I just had to create pictures of you know pandas on an airplane, pandas playing baseball. And I was like, wow, this is really cool. AI can create pictures, that's great. Um

George Danilovics

But now I go to Claude for everything. Um, if I've got a question about a technology, I ask Claude. If I want to know a recipe, I ask Claude. If I'm curious about how to build something, I ask Claude. Whereas before I used to go to Google, I used to go to a search engine, I used to go to a vendor's page. So that just happened over time gradually, and I never noticed a behavior change until after it happened. You know, where I used to go to the Google search button on my phone, I now click the Claude app. Um,

George Danilovics

So that systemic piece, it's hard to see it as it's happening. When you look back, you'll realize when it happened.

Mimi Yeh

Um, I'll add a little bit onto this, and then I know we've got so many more questions and comments to parse through.

Mimi Yeh

I think it's absolutely true that tools change. I am old enough to remember when Yahoo and Ask Jeeves and all these other engines and methods of getting information, um, we used tools that do not exist anymore today. Or if they do, they're probably all corralled up in Bending Spoon or one of those other companies that have smooshed together all of those cats and dogs.

Mimi Yeh

So the tools change, but why you're using the tool and why your organization chose to apply AI to achieve a particular part of your strategy, your mission, or your goals, that didn't really change.

Mimi Yeh

So I guess one thing I would say is start with your strategy. Start with what is the opportunity or the challenge that you want AI to solve or assist you with. Um, and on those same points, maybe a side tip is to document that journey. Because if it migrates from one AI tool to the other, the logic and the thought process that you used probably will not change. That will stay sound and having that document might make it easier to feed those prompts or to feed that, create that Gem that is going to be customized to your organization with the proper prompts and the way that you want that AI persona to uh behave when it's interacting with your staff. Um so you know, document that journey and be clear about why you made the choices that you made.

Carolyn Woodard

Those are so such great pieces of advice, Mimi. Thank you so much. Um,

Carolyn Woodard

We're getting close to the end. I'm gonna open a huge can of worms because we are close to the end. Um, Greg asked this really brilliant question in the QA that I saved to the end because I knew we like this could be an hour-long discussion, basically. Um,

Carolyn Woodard

But I wondered if, kind of as closing thoughts, uh, we've started in this webinar asking about like maturity. So you're already using AI. How do you get to that next phase where it's kind of really integrated?

Carolyn Woodard

And Greg asks, how do you address and demonstrate care for genuine concerns about ethics, social impact, all of the issues that we know about AI and that nonprofits care about around AI while you're trying to put this formalized and mature adoption in place? So he says, many of the people in my sector are opposed to AI based on economic and ethical concerns. I don't want to ignore that. And I want to ensure that we move forward at a pace that is good for our community and, of course, for your organization and then for yourself as well, like what you feel comfortable with.

Carolyn Woodard

So when you run into that, these sentiments at clients and when you're working on AI and with your colleagues, do you have maybe some prompts for us of ways to think about it? Since this is a very long conversation, we only have a few minutes, but how do you start answering that question? I ask you a really, really easy one.

George Danilovics

Easy question, yeah. And I've heard this, you know, this concern, you know, from from multiple individuals and organizations as it pertains to AI, both on the social piece, the the concerns over uh energy and what it's doing to our environment. Um and

George Danilovics

I think all of those are valid concerns.

George Danilovics

I think it is something that we as users and stewards of AI should be looking at other organizations that are trying to make sure AI is used as a uh a force for good. Um and that, you know, as it's using resources, we use it in a productive manner. And it's not used to the detriment of society. Um

George Danilovics

I don't think any one individual or one group can sway an entire technology. Um, but each organization does have to think about how they are best going to use a technology or not. Um, and some of those conversations may require, you know, consulting with your board of directors, um, particularly around, you know, if there's there's concerns on the social piece or energy or resources piece or the ethical side of you know, where an organization has said, you know, where are those red lines of where we're not going to use AI. Um and

George Danilovics

To have that as a board decision, not just a staff decision. Um, and so I think it is something we have to look at holistically and we we have to be cognizant of some of the the trade-offs and some of the consequences of the technology.

Mimi Yeh

Yeah, I think that's right. And I I would say, you know, I think we can all agree that um AI every major technology raises new ethical questions and dilemmas with it. AI is not different in that case. It may be different in what the ethical questions are and the degree of the concern, but you know.

Mimi Yeh

I think a lot of this what I think about is the answer to whether and how to use AI should be reflective of your overall technology strategy and your technology approach because there's lots of other considerations and concerns and ethical dilemmas with many other technical tools that are out there. There's all kinds of concerns associated with flying and using a plane instead of other modes of transportation. There's ethical concerns associated with using tools like Salesforce because of the way that their content is being used by certain actors and players out there.

Mimi Yeh

So we may not be able to eliminate all of the risk. We might have to instead manage it thoughtfully. And to what George said, be transparent about how you choose to use it and make sure that the use of it still aligns back with your mission and your values. And you're still operating in a way that you're protecting sensitive data and respecting people's privacy and being cognizant of potential bias that might exist within different AI tools and how you as an organization elect to proactively manage those.

Carolyn Woodard

I love both of those answers. Thank you for weighing in with on something that is such a serious topic. That really helped.

Carolyn Woodard

I think I just want to echo what you said about transparency. We've been really recommending that it is an all-staff ongoing conversation, like also with your executive team, also with your board, because you can find that you have staff that like you have a policy, but they're not okay with it. So you really need to be constantly having this evolving conversation about your values and how this relates to what your values are.

Carolyn Woodard

And another thing I just wanted to say that's I've been thinking about recently is especially around like environmental issues, looking to like the big environmental organizations and what they're doing. And many of them have put out like how to think about AI, how to use AI, and you know, kind of where they're coming to it from. They have a lot more experience than I do in terms of what it's doing to the environment. So kind of hooking on to people in our peer group that have more experience than we do is another way to address those fears and concerns. Um,

Carolyn Woodard

So I want to move on. Uh, we only have a minute or two left. So I wanted to, I think we hit the learning objectives really well. Thank you both, George and Mimi. Uh, we wanted to talk about it, the AI maturity model, and give you some clues of uh how to figure out where your nonprofit fits in there. Um, how do you build AI as a standard core competency across your workforce? We talked about the training and making it boring and everywhere. Um, and then what does strategic AI implementation look like and some of these best practices around, you know, training, um, measuring, thinking about this maturity model.

Carolyn Woodard

And I love the idea that you don't have to get to the top of it, but just figuring out where you are on that model will help you with where you want to get to as well.

Carolyn Woodard

So um I want to tell you all about the webinar that we have next month. Uh, if we invite you back for that, I'm gonna be leading that presentation with my colleague Eric Solce, who uh is helping me on a project that I'm doing right now. So we're gonna report back on how it's going. Uh,

Carolyn Woodard

I have an ongoing project using AI to help build a data lake for marketing. And we're gonna talk about the structure that we've come up with that helped me manage this giant project with a little tech coaching, a lot of AI. We're gonna share how the project evolved, what didn't go as well as we hoped, what we learned. Uh, we think that kind of a structure has a lot of potential for nonprofit AI projects.

Carolyn Woodard

So if you have been putting something off because you don't have the technical know-how to do it, or it's just this giant thing that you don't have time for, I know some of us put in the chat, like, what would you do if you could do it? Um, so if you're thinking about that, um it uh whether it's you know building a data lake for some other purpose or analyzing years of program reports or you know, whatever it is that you've just been putting off and want to get started on, that's what we're gonna talk about next month. Um,

Carolyn Woodard

That is at 3 p.m. Eastern Noon Pacific on Wednesday, August 19th. I'm gonna share that link with you in the chat. Uh, it's gonna be on our website, community it.com.

Carolyn Woodard

And then please join us on Reddit at r/ nonprofit IT management for another 30 minutes or so. We'll be uh answering your questions that we didn't get to that were here in chat. Um, and oh yes, I'm sorry, I put I put the specific uh link in that goes exactly to the QA, but you can also find us on r slash nonprofit IT management. And um,

Carolyn Woodard

Thank you so, so much for joining us, everyone. And an hour of your time is a gift, and we really appreciate it.

Carolyn Woodard

Thank you for spending this afternoon time with us, George and Mimi. I just appreciate your time as well and all the thought and effort that went into this presentation. It really helped me. I change a little bit, reframe how I'm thinking about this also. Um, so thank you again so much for coming.

George Danilovics

Thank you.

Mimi Yeh

Thank you all. Have a great day. Thanks.

Community IT Intro

Thank you for joining this Community IT podcast, part two. You can find part one in your podcast feed if you subscribe wherever you listen to podcasts.