Name: AI Driven Nonprofits: Harnessing Technology Without Losing Your Heart
Uploaded: 2026-09-22T18:32:05.000Z
Duration: 1 h
Description: AI Driven Nonprofits: Harnessing Technology Without Losing Your Heart

Transcript for "AI Driven Nonprofits: Harnessing Technology Without Losing Your Heart": Welcome, everyone. Thank you so much for joining us today for AI Driven Nonprofits, Harnessing Technology Without Losing Your Heart, presented by one of our favorite nonprofit experts, Steve McLaughlin. So before we kick it off and I pass it over to Steve, I just wanna go over a few quick items. You'll receive a link to the recording as always within about twenty four hours to the email address that you registered with. And please feel free to enter questions as you think of them throughout the session into the q and a. You'll find that there is an interactive chat, so please feel free to, engage with us throughout the session, connect with your fellow peers. And then don't forget that you can access incredible resources to help navigate AI in your mission's next chapter in our doc section as well. We also will be using some fun polls today. So we'll be sharing those on the screen when they come up, and they'll also appear next to the chat and messages tab, over to the right. So just to keep that in mind. So without further ado, I'm gonna go ahead and pass it over to our expert today to dive on in. Great. Thanks so much, Abby, and thanks for everyone joining us today. I'm Steve McLaughlin, and I wanna talk about AI, artificial intelligence. So to start, let me pose a potentially controversial statement, which is your biggest AI problem is not AI. I've spent a number of years working on different types and forms of artificial intelligence, working with nonprofits on their use of AI, and continually, I keep coming back to this statement, which is, you may have challenges and problems in the adoption and use of AI, but in a lot of cases, you and your organizations, your biggest AI problem is not, artificial intelligence. It's other things, and we're gonna try and spend some time today talking about those other things. So as a bit of context, how do we get here? Well, about ten years ago, I wrote a book called data to different nonprofits. And one of the things I found in that book was that a significant driver or barrier to nonprofit organizations using data in more effective ways was culture and strategy and how they thought about things, that it wasn't necessarily about the data. Fast forward ten years, and I wrote another book that came out earlier this year called AI driven nonprofits. And in many ways, it's sort of the sequel to data driven nonprofits. And there must be something good in there because, it did spend a good amount of time in the number one bestseller category for nonprofits on Amazon. And so I'm gonna share some of the insights, and things I found as a part of the book, but we're also gonna get a bit interactive here today. And so as Abby noted, we're gonna have some, polls and things like that to try and engage on some of this content as we explore. What does it mean to be an AI driven nonprofit? My one caveat, word of warning is I am not gonna spend a lot of time talking about the tools. We could do that. Maybe we have another webinar that's all about the tools, but, like, there are lots of tools for AI, but there's a tendency if you just fixate on the tools, sort of the old adage, if all you have is a hammer, everything looks like an ale. Yes. We may talk about examples in terms of the use of tools, but if you're expecting today to be about how do I use Check UPT or Perplexity or Google Gemini or Cloud Code to do stuff, I'm probably not gonna spend a lot of time there. But certainly as we get into questions and things like that, I'm happy to share firsthand examples of some of the tools. But I I'm a big believer that strategy comes before, technology. And so we're gonna talk about some strategy stuff and some culture stuff and all those kinds of things. So to get us started, I wanna talk about butterflies. So today's talk is gonna start out, and we're gonna talk about butterflies. This is a quote that's in the book, and I love this quote from Buckminster Fuller, which is, there's nothing in a caterpillar that tells you it's gonna be a butterfly. And I feel like there are a lot of us as individuals and as organizations right now who are caterpillars somewhere in this transformation to becoming butterflies or AI driven butterflies. And so what I always find helpful whenever there's a new trend, there's a new technology, there's something, that we need to adapt to that's gonna cause some transformation. We should recognize that that's not always easy and it's not always obvious, but it continues to happen, to both us as individuals and as leaders in our organizations. And so it's okay that maybe we don't have all the answers today. We don't necessarily have it all figured out. Hopefully, what I'm able to help you today is to get yourselves ready a bit more about being more AI driven, your organization's more AI driven. And at the very end of today's session, I'm gonna give you a thirty day road map for how to, drive to being much more, AI driven within your organizations. So we're all caterpillars right now on this journey. That's okay. So first thing I want you to think about is I want you to think about what is your AI usage, you as an individual. Right? And there's three different sort of levels of AI usage. High. I use it daily or multiple times a day. Right? I'm actively using it every single day in the work that I'm doing. That would be considered high usage. Next is medium. One to three times a week. I don't use it every day. I do use it multiple times per week. It sort of depends on where I'm at and what I'm working on at the particular moment. That would be considered medium. And then there's low. Never or very rarely, not very frequently. I don't really deal with it a whole lot. That would be considered low. So have a think for a second here about how you would categorize your AI usage. And then Abby's gonna help us here. We're gonna put a poll up, and I want everybody to respond and say, is your AI usage high, medium, or low? So I'll pause for a moment. I'm gonna take a drink, and we'll see all of these votes come in. Remember, vote on the screen, not just in the chat. Great. Lots of votes, lots of participation. It's always interesting every time I do this to see what the what the distribution is here in groups. Alright. Lots of votes in here. Lots of votes in here. So looks like it's an interesting split about 34%, 33, it's moving on. 34% of you would say high. About 42% of you would say medium, and about 22% of you would say low. And again, get in where you fit in. Just I'm trying to understand where is this audience today in their usage. So this is really great. Okay. Alright. So that's usage. The next thing that I wanna talk about is what is your AI openness? So what do I mean by openness? So, hi, I'm a believer. I think AI is gonna make us more effective. We're gonna drive things a lot faster. I'm energized. I mean, you're I'm high. I'm all in. Right? Medium. I'm cautiously optimistic. I see the potential, but I want some guardrails, and I need to have some proof that it works. So I'm balanced, I'm curious, but I'm measured. And then there's low openness to AI. I'm skeptical. I worry about accuracy, bias, what it means for jobs, what it means for what we're doing. I've got some anxiety here. Right? So think about where do you fall on your AI openness. And, again, Abby's gonna put up a poll here, and some of you are already voting. Give some time for some votes. Lot of mediums. If if we were to a sporting event and this is on the big jumbotron, we'd see, you know, mediums winning, mediums in the lead. Give you guys a few more seconds here. Great. So, again, high openness, about 21%. Okay? 57, almost 58% would say you're at a medium, and then low, about 21%. So what's interesting there is highs and lows are almost about the same percentage. And someone in the group says, I'm all of the above. We should talk about that. Can you really be you know, some days, it's it you know, I'll I'll be fair. Some days, the day of the week, my openness does change depending on the day of the week. So that's great. Okay. So I asked you about your, usage, and I asked you about your openness. And you may be wondering, well, why did we do this, Steve? Okay. Here's why. I want you to think about how you ranked your openness and how you ranked your AI usage, and I want you to find on the grid your AI spirit animal. Right? Are you an owl? Are you a chameleon? Are you a wolf? Are you an eagle? Are you a squirrel? And we didn't put a poll in here because I thought we'd have fun. Drop in the chat. Tell me, what's your what's your spirit animal? How are you feeling today? And, also, do you do you think that really is your spirit animal? Got some owls. Great. Lots of foxes, eagles, foxes. Yep. The probably the reality is we're gonna get there's a bias here. You're here because you're interested in AI. You're trying to learn more about it. We we may not have a lot of ostriches in here today. Got some rabbits. Right? It's not a snail in the middle. It's a chameleon. It's a chameleon. Yeah. Foxes, champions, ostrich emu. Okay. Love it. Okay. So that so that's where we are. And, also, you know, some days you might feel like a fox, and some days you feel like a squirrel or a chameleon. You're not always gonna be the same every single day, but I think it's important for you to sort of you as an individual, not your organizations. Think about where you are, where you fall in here because, that that's important as you think about your own sort of adoption and use of AI. So we'll do this exercise. So so you're you have your spirit animal. So so who cares? So what? Who cares? Right? Steve, what do I do with this? So one thing I want you to think about is what are things you could stop doing, things you could start doing, or things you might redesign than what you do today? So let me talk through a couple of examples. So things that you could stop doing because AI might be, helpful to you. So where am I working harder than the task requires? Right? So it could be things like, do you retype meeting notes by hand afterwards? Think about manual tasks, manual things that you do. Are there things where AI could potentially help you stop doing that manual work, those types of things, and it could help you in some other areas. So think about meeting notes, routine reports that you do line by line on a spreadsheet or a pivot table, preparing internal reports, those kinds of things. Could AI help you in some of those ways? So think about what you could stop doing. Then I want you to think next about what could you start doing that you don't necessarily do today? So what would make the biggest difference if I started tomorrow? Right? So examples. I use AI to draft meeting summaries and follow ups, and certainly a lot of tools today, Teams, Zoom, Slack, etcetera, etcetera, have a lot of this built in where you can get a meeting summary as, it tells you what the action items are or follow-up items. Right? Could I get back in some cases, could I get time back by start using this? Or I know for a lot of us, it's I don't do this today because I gotta move on to the 15 other things on my plate. So does it help me start doing something that I'm not actually doing today? For frontline fundraisers, could I ask AI to help, prepare a briefing before I have a donor call? Right? I'm about to have a a meeting or a call with a mid level or a major gift prospect. Can I have AI prepare a briefing before I have that actual call? Right? I don't do that today. I asked someone else to do that. Right? Could I ask AI to help me with this? Right? And then one of the things, you know, is there a new tool or something you could try to take on, learn on purpose, not accidentally. Oh, I accidentally started using this, that you could start doing this month potentially you hadn't done before. And then the last area, that I want you to think about is something you could redesign. So what would I do differently if I started from scratch today? You know, the classic, we've always done it this way. We've done this for years and years and years, but with the use of AI, I might, again, I might think about how I prepare for a board meeting, how I prepare for an internal meeting, how I prepare for a conversation with a gift prospect, how I write or revise, grant applications, email content, all those kinds of things that you do. Really start to think about, is there something that you might potentially be able to redesign that's different than how you've always done it because you've got AI. And I'll pause for a second. Have a think here and just throw in the chat something you think you might stop doing because AI could help you, something you potentially could start doing because of AI, or something you might redesign because of AI. And, again, we're all friends here. Don't be shy. Drop something in the chat. Love to hear sort of something that just immediately comes to mind that you could potentially doing. Love it. Board minutes. Yeah. Redesigning newsletters, updating portfolios. There's not a shortage of stuff. See, this is the whole point. Stop making PowerPoints. I love making PowerPoints slides, but that's just me. Yeah. Lots of ideas in here. And this is where it starts. Right? And so back to, you know, your spirit animal, where you are today, you might be different in the future. Really encourage you, use this approach. What can I stop doing, start doing, and redesign? It doesn't have to be everything. Right? Even if it was just one thing this week, tomorrow, before the end of the week that you started doing differently. Love it. Lots of great ideas. AI for donor briefs, personalized thank you letters, redesign gift entry. Man, that should be the next webinar we do. How how would we redesign gift entry in the world of AI? Love that one. Okay. So so far, we've been talking about all of you, your butterflies, and where you are at individually. Gonna shift gears here in a second, and we're gonna talk about where your organizations are because each of you are at an organization. And since I can't ask the organization questions because it's not a thing, you are. We wanna shift gears to, the organization side. So, we were talking about butterflies. Now we're gonna talk about robots. Okay. This is a quote I love from Larry Tesler, which is AI is whatever hasn't been done yet. So AI has been around a long time. Roots of AI go back to the nineteen fifties. We're not here for a history lesson, but the thing's been around a long time. Okay? It's just that what tends to happen with artificial intelligence is when we, give it a name, we move on. So for example, I have worked with, predictive models and machine learning for well over a decade, and we could have people on this call who would say, but, Steve, machine learning and predictive models, that's not AI. Yes. It is. It's just a subset. And you but usually what happens with it is once we give it a new name, it stops becoming AI. So I love this quote by Larry, which is AI is whatever hasn't been done yet. But in fact, there are real definitions of real things that, are done with AI. So, couple things, just the level set here. What what AI actually is? Okay. For starters, it's not one technology. It's not one thing. It's a category, not a single tool. Just like, you know, if I called something a vehicle, we're talking about, bicycles, scooters, jet airplanes. They're all technically vehicles. Right? It could be a predictive model, a chatbot. Right? They don't necessarily do the same thing. The other way I like to think about AI is it's a ladder, not a switch. It's usually not an on off. Right? It's where is your organization today, where is it tomorrow, where is it next week, and how do you continue to mature in your use over time? The myth would be, oh, you know, can't we just flip a switch or turn on this setting and boom, we're AI driven. No. It doesn't really work that way. It's climbing a ladder of skills and behaviors and cultural things. And the other way it's like to also think about AI is AI is being an extension, not a replacement. How could it help me, amplify what I do? And let's keep in mind, AI amplifies good things and AI amplifies bad things. Right? And, you know, word of caution. Right? If you have a bad process, AI just makes the process worse using AI. So a lot of things organizations are running into right now where they're struggling with adoption and pilots and this and that, it's because oftentimes, like, the core process they're using, it's a bad process. And so if I take a bad process and apply AI to it, I I just get a bad process with AI. Right? And so we need to sort of think about it within that context. Now, in general, most people would agree that there are six types of AI so far. Eventually, they'll invent seven eight nine ten. Right? So I wanna talk about these, and then we'll have another poll question. So, predictive AI, so things like, a predictive model for fraud detection, lapse donor detection, matrix likelihood, a predictive model. Right? It's attempting to predict the likelihood of a future outcome by analyzing data. Predictive AI, been around quite a while. Generative AI. Right? So this is all about content. It creates new content, text, images, code, videos. So your chat GPTs, your mid journey, your gamma. Right? It creates content based on data and other types of things. Right? Agentic AI, it's one of those buzzwords these days. So Agentic AI is just how does something, how does AI help perform a multi step workflow? There's multiple tasks. They come in a certain order. Agentic AI can go do those things, both, on its own, but also automated, which makes it very powerful, where it can go off and complete a series of tasks and, you know, crosses off the to do list. Right? The top three on the that appear very common. Then but there's some other types of AI that nonprofits engage with as well. So physical or embodied AI, so think drones, self driving cars, warehouse robots, you know, where you're getting into the use of vision, sensors, robotics to take on certain tasks. Conversational AI. Right? So things like chatbots, virtual gift officers, digital agents, right, where there is, voice and text, and those are engaging with people in a way that's sort of normal human natural interaction. Right? And then there's assistive AI. So things like, autocomplete email sorting, which we don't probably you probably even think about this anymore. Like, when I'm on my phone or I'm in my email and it tries to autocomplete a sentence, that is a type of AI. It's called assistive AI. And, again, we don't really we probably don't even think about that as AI because it's so commonplace, and yet that is a real thing that exists today. So I think, Abby, our next poll is to ask the group what of these types of AI, what are the types that you your organization is using today? So very curious what you're using or maybe what you think that you're using. So we got some predictive, some generative. Yep. Lots of people are using generative AI. Not while this presentation is going on. Maybe you've got some Magenta going on right now while you're listening to me. Some conversational. Haven't seen any physical, some orgs, depending on what your mission and focus is. That does come up. Right? So a lot of use of generative AI, some assistive AI. The thing about predictive AI, I am willing to bet that you may you may be benefiting from the use of predictive AI. You just don't you just don't know, that it's there. One question about Office three sixty five. Yep. Oh, there's a lot of assistive AI built into Office three sixty five. Yep. Absolutely. Great. So that just gives us a sense of where your organizations are, which is good. Now I'm gonna jump back to my slides. Okay. So that was the types of AI. Now we're gonna talk about a where you are in, in maturity. Right? So where where is your organization in this transformation from being a caterpillar to a butterfly? So talk about a couple of these. One, the first step in the maturity curve is you're aware. Right? It's in the use of AI at your organization, it's it's informal. There's no policy. There's no standards. People might be using it here and there, but it's largely being driven by individuals. That's okay. Second step in the maturity curve here is active. So, you got some structure that's starting to take place. It's becoming more coordinated. Right? Maybe you have a draft policy. There's some repeatable use cases, but it's we've moved on from sort of wild wild west to, starting to take shape in the wild wild west. Stage three, operational. So you've got some governance that's coming into place. There's some clear roles, real policies. People feel more confident in their use. They're not unsure about, oh, is this okay to use? Is this not okay to use? Right? That tends to happen when you're in stage three operational. Stage four, systemic. Right? So AI is built into the core workflows. It's not a side hustle. Right? When we are doing a particular program, a process, something we're doing, we built AI into a part of that workflow. And so how does that feel? More stable, more repeatable. It's more widely accepted, those types of things. It's part of what you're doing. And then stage five, this is the hard one to get to, transformational. So AI is intentionally aligned to mission outcomes and strategy. How does that feel? Aligned, purpose driven. We're not all eagles and foxes. That's okay. But as an organization, you know, we are really doing things from a transformational perspective. And this is a hard one to get to. So we're gonna ask, where is your organization on this continuum? So aware, active, operational, systemic, or transformational. Very curious where everyone is. And, again, it's always fascinating different groups either in person or on a webinar like this. Always interesting to see where where the group feels they are today. So sounds like a lot of some aware, active. Done this multiple times. Little secret, no one in the room has ever said they were transformational. So not surprised to not see transformational show up here because, look, this is still new in a lot of ways. Right? So sounds like it it's a good mix of a lot of orgs orgs feel active, moving towards operational. A few two out of you with us today in the crowd. Feel like you're systemic, which is great. Right? And so I just one thing I would just encourage you to think about, this is where you are today. You could you could play the same, stop, start, redesign exercise for your organization, as a way to think about how if we're active, what would we have to do in the next twelve months to become operational? Right? If we're operational, how do we move to systemic? It's a journey. We're all on that journey. Just think about where is your organization today? What are some of the things that you could potentially do to level up on the maturity scale? Okay. Alright. So we talked about butterflies. We talked about robots. We're gonna talk about stoplights. So like this quote from Alice Walker, look closely at the present you are constructing. It should look like the future you are dreaming. So I wanna dive here into a game we're gonna play. We're gonna play a game called red light, yellow light, green light. It's not squid game. Right? So it's red light, yellow light, green light. And I like to play this game because this is really a game secretly, this is a game about governance. But we know organizations struggle with governance and, oh, we need to form a committee, and it sounds like a big deal and stuff. And so, we're gonna play this game to get you more comfortable with thinking about governance. So I have a bunch of scenarios. Some of them they're almost all real. They're certainly real and organizations are doing them. The question is, do you feel that this this scenario should be a red light, a yellow light, or a green light? So red light means stop. This needs a policy conversation before it happens again. Yellow, proceed only with guardrails. We need to review it. We may need to put some limits in place, and then there's green light. This is good. Keep going. This is safe. It's appropriate. It conforms to our policies and standards. Great job. Keep moving. Alright. So I'm gonna show a scenario. I'm gonna read through it, and then go ahead and drop in the chat whether you think it is red light, yellow light, green light. K. So the first one, scenario number one. A predictive model flags 40 lapsed donors most likely to reengage if contacted this month. Two gift officers review the list and decide which donors they'll personally call. No autonomous outreach happens. Every call and email is written and sent by a human person. The model's recommendations are logged so the team can check its accuracy each quarter. What do you think about this one? Red light, yellow light, green light? Lots of greens. I'm always curious to see on this one if anyone thinks it's not green. Oh, got okay. One or two folks think it might be yellow. I just saw a red in there. Okay. For folks who so I I think most the majority here would sort of thinks green. For those of you who think this might be yellow, just drop in the chat real quick. I think this might be a yellow because why? Very curious. And I'll I'll try and share that. I think we can all see the chat here too. So some of you think it might be a yellow. Okay. Might be yellow because it's the use of donor data. So right? So do we have a certain policy in place? Who gave permission? Okay. Great. Always check on that. Right? That because you are dealing with with data, it could have PII. You know, this is a four bullet point scenario. Some details were left out. So yeah. So for some orgs, you might say this is a yellow because we wanna make sure where did this data come from, how is it being handled. Totally good. Love that one. And then some folks, said red. I'm gonna assume for similar reasons, which is, does this, it may go against the policy that we've got in place, who has access to it, etcetera, etcetera, etcetera. Great. Love that one. Okay. So that's scenario one. These do get a little bit tricky, or at least I tried to make them get tricky. Alright. Scenario number two. A staff member uses generative AI to draft personalized thank you letters for 2,000 year end donors. Each letter references the donor's giving history and preferred program area based on CRM data. The staff member skims a sample of 20 of the 2,000 letters. Rather than reading every one, donors receive their letter faster than the old manual way. So I have to think. Red, yellow, green. Reds, reds, reds, yellows, yellows. I'm waiting to see if there's a if there's a green. No. So far, nobody said this one is a green. Red. Red. Red. Red. Red. Okay. So the reds and yellows, I think I know why. But tell me from your perspective, what what's the what's the red flag here or the yellow flag here? Curious. K. Not enough oversight, not enough QA, PII. Right? So there's common threads here. Right? Like, okay. PAI, PHI, which, I deal a lot with the health care side, happens a lot. Right? They looked at 20, not 2,000. Right? Now hint where I'm going here. Is any of this stuff AI stuff, or is this just regular data stuff that we've gotta pay attention to? The QA, who has access to PII. Right? Quality control, all those kinds of things. Great. Okay. This is always a always a fun one to do, to see what people think. Alright. Scenario number three. A program director uses a free AI writing tool to draft a grant report. Unpublished program outcome data is pasted in to make the writing more specific. No one has reviewed the tool's data privacy or retention policy. I'm already getting the stops. The final report is submitted without disclosing that AI was used. Alright. So read read read read read read read. But you see the point, like and I haven't seen anybody. Yeah. I haven't seen anybody bright red on a couple. I forgot one of the icons is a red. I should remember that for next time. Yeah. So in most cases, this is like a big there's so many things here that are on fire and you shouldn't do, and yet we know people wanna experiment. People wanna try out things. And so, again, if you don't have policies in place, depending on where you are in the maturity curve, you can see how you can break into jail really quickly here by people have a well intentioned, but oftentimes do things that, are a red light. Okay. Scenario number four. A gift officer connects an AI note taker to video calls with major donors. The tool automatically transcribes every conversation, including sensitive personal and financial details. Its default settings share meeting summaries with a third party vendor for product improvement. No donor has been told their conversations are being recorded or processed or processed for AIs. So, again, lots of reds here. Reds, red, reds. Now let me pose this a slightly different way. What would you have to do to get this from red to yellow or from yellow to green? Talk to me. What would you require, to move this from a red to a yellow or a yellow to a green? Okay. So disclosure, inform donors, asking for permission. Yep. And that's and especially with I mean, a lot of this stuff gets recorded these days. A lot of times, there's a prompt in the software. Whenever I'm talking to, clients, users, things like that, I always ask permission to record the call even before it starts. Yep. Disable the product improvement. Maybe not you don't want to disable product improvement. You may wanna disable that it shares certain certain data. Yep. So that's another way to think about these scenarios, which is, yeah, wow, red, not happening. But there's some good stuff in here. Right? Okay. I like that a gift officer would wanna get notes from a conversation, but how do I do that in a way that is yellow or green that does not, cause problems. Right? So good idea, but we need to be careful about how we actually do it. Okay. The last one here, I've only had five of these today. Fifth one, an AI tool analyzes internal and external donor data and recommends how to segment and appeal. Generative AI drafts the direct mail and email copy for each segment and message to about 5,000 donors. A staff member has the option to review and edits every version before it goes out. The appeal raises 15% more than last year, but donors were not told that AI was used. So what what do we think here? Red, yellow, green? Some of you have said red to every single one of these. Some of you are yellow. Do we need to disclose? Do we need to tell donors? Great comment. I'm conflicted on how much to disclose to donors. We might wanna talk about that one. Yeah. Okay. Some yellow. Some of you think green as long as the policy was reviewed and followed. And so, again, this isn't always an easy, you know, easy decision. Right? There are some parts of this. Okay. I got a better result. They followed a policy. Things had to be reviewed. This disclosure thing comes up. It was interesting. I was I was doing an in person version of this last week, and someone came up afterwards and said, you know, it's interesting. There's all this talk about disclosing that we use AI, and it is interesting. She made the point to me that I thought was was pretty insightful, which is she said, you know, I don't tell donors or people we engage with that I used a pivot table or that I used, certain kinds of technology. I didn't use Excel. It's inter it's like that's an interesting point. I think in the and my response at the time was, I think just where we are today, we wanna make sure that, if there's anxiety, if there's apprehension, if there is, an abundance of out of an abundance of caution, these might be things that we want to check. And maybe that's something we do need to do more of today is disclosure, until in the future, we don't need to do it anymore because it's sort of just known. But I thought it was that was an interesting question to make. Okay. Alright. Next, we did stop lights. So, again, just one way to think about these, something that might be a stop today, red light. Think about, okay, what could this actually help us with? What would what would rules we could put in place to make it safe or safer to try? Right? And as we do more of that, what have we learned that scales? Right? Things like look. We have a policy that there's always a human in the loop, and that's not optional, and you can't, you know, pass go and collect your monopoly money. Right? You've gotta have a human in the loop on certain types of things. And you might also decide there are certain things where automation, is okay, but, again, you might be on a different journey. The way to sort of think about this, though, is how do I move from no to how? Again, what could we do different? What could be a rule we put in place? How do we get more comfortable, with doing all of these things? Love it. Okay. Part four, choices. Jim Rohn quote, the future is not created by chance. It is created by choice. So we're gonna get into some org culture stuff here. And the way I like to frame it up is everything's a choice. Everything's a choice. Do we wanna do it? Do we not wanna do it? How does it fit in? And and people often struggle with making decisions or prioritizing decisions. That's a very human thing, and AI hasn't solved all that for us yet. So when I'm thinking about the use of AI, you'll see this is probably the third time today where I have an axis of two different things we're looking at things. So in this case, if I'm thinking about what we should or maybe shouldn't do or try or not try with AI, I think about in terms of impact. Does it have high impact or low impact? And what's the risk? Is it low risk or high risk? So love ideas. Ideas, scenarios, they could come from anywhere, a staff meeting, a suggestion box, a, a one on one, something came up. Great. Take those and plot it on this grid. Right? Is it high impact, low impact, high risk, low risk? That helps to give you some guidance on what you potentially do. Right? So easy place to start. Things that have high impact and are low risk, those are quick wins. Right? And so if you believe it's high impact and relatively low risk, then how do we do that? How do we put a policy in place? How do we encourage people to do that? So things like, major donor meeting briefings, thank you call scripts, summarized clean donor data trends, directly mail layout variations. Right? There might be things you'd be able to do that have a high impact, relatively low risk. Those are quick wins. We should try to do those. Or when we get ideas or we get suggestions around things we could potentially do, I'm gonna put it in the quick win category. Alright? Box one over to the right, two, strategic moves. So these things that are high risk, high impact. So there's a balance here. Right? This could be game changing infrastructure and the use of AI that really transforms what you're doing. But but but but, it requires strict governance and planning. So it's high risk, but it's also high impact, high reward. Right? So how are we using, you know, AI assisted prospect research, automated content personalization, virtual gift officers, all those things could be high risk, high impact depending on how your organization thinks about it. Those are things you should consider. You just need to know how you wanna think about them. Then there's the bottom, left hand category, what I call fillers. If I'm being nice, they're fillers. If I'm not being nice, they're distractions. They're distractions disguised as quick wins or strategic moves. So it's low risk, it's low impact. So there's not a lot of risk, but it also doesn't move the needle. Right? Easy to do, but it does not add a lot of value. Now if previously, previously on Steve's talk, if you said your organization isn't as mature, isn't using a lot of AI, then even having some ideas, some things you potentially do that are totally fillers, it's good. It's fine. Right? It may not be high impact, but it is low risk. And if that's the thing that gets others in your organization to move and do some stuff, then that would be a thing you should think about and consider. And then, obviously, the bottom right hand, the danger zone, high risk, low impact, don't do it. Right? The unvetted tool, the things that have potential legal liability, PII, things from that scenario as we just talked about. Again, when ideas and things come up, drop it into, this one of these particular categories. And be clear of, like, hey. Love the idea. We just right now think that's high risk, low impact. So we're gonna put that one in the danger zone. Maybe we revisit it over time. Right? And you guys can probably think of a bunch of ideas. In fact, someone just dropped one in the chat. Where do email tone checks come in? So lot of AI stuff these days is your ability to, adjust and tune the tone of a message. Now I would argue we we could do this before AI. It's just AI helps us do it faster and at more scale. Right? I would say that email tone checks could be a one or a two depending on where your org is right now. With something like that, right, always test. Right? Hey. Let's try this with an initial group to start with, and then maybe we roll it out. A lot of this stuff, I don't recommend the big bang approach. Tomorrow, we're gonna flip the switch and suddenly everything works this way. But yeah. So it could be you know, it might be something you wanna test. Hey. Low it's how do I decrease risk? It might be I test it with a smaller group of people. Let's find out. Does it have an actual high impact? And then you might move it from, you know, low risk, to high impact or or something like that depending on what you're doing. So, again, I I think great idea. Just try it, test it out, see what happens. Okay. So just general recommendations. Use this grid. You can do this, back at your organization. Right? Is if something's a quick win, go. Right? You don't need a committee. You don't need a whole group. Empower people to do it. Maybe you need some policies in place to clarify what something is a go, but encourage that. Things that might be a strategic move. Right? Because there's higher risk involved, we might wanna pilot it. Right? We might need a data or a privacy policy review. We might wanna pilot it or test it on a particular group. Right? But think of it as a project, not just an experiment. Hey. Let's find out what happens. No. Let's treat it as a project. Let's staff it appropriately, time box it. I think I think the thing about pilots, they get into trouble as they seem to be never ending. So is this a thirty day pilot, or is it a ninety day pilot? Right? Time box it. Fillers and distractions, use your judgment. Right? Especially right now where, it may help your organization to drive some adoption and use. You're trying to get people to to use some things. You know, use your judgment, but but fillers might be helpful now. And then, obviously, you find things that are in danger zone. If you're in a danger zone, stop. If you find yourself digging a hole, the first step is to stop digging. So stop if you find yourself in the danger zone. Okay. We're gonna wrap up in the afternoon. Well, it's afternoon for lots of us, but not everyone. We're gonna wrap up with breakfast. So this is a quote from Peter Drucker, culture eats strategy for breakfast, and then AI is gonna come along and eat your lunch. So wanna address, a couple of real concerns that organizations are facing these days. There are some AI culture challenges that don't have a whole lot to do with the technology or what tool you choose to not use, and they're real. And I wanna point and these are just six of them. There are certainly others we could probably talk about as well. So I just wanna note these as things to think about and be aware of in your organization. Right? So one particular cultural challenge that we have to deal with with AI is deskilling. Right? What happens when AI does the thinking for us? Right? I used to always do this. I was pretty darn good at it, but I've turned that over to AI. Do my skills atrophy because I'm I'm not I'm not doing it anymore. Right? That's a real cultural challenge. Trust. When should we believe the machine or not believe the machine? Security. Whenever you're dealing with data, technology, super important. Right? What happens when our AI data become when our data becomes AI's input? I deal with this probably every single day in the work that I do. Lots of questions about how data is used and not used and those kinds of things goes back to just because you can't do something doesn't mean you should or can. Shadow AI, this is an interesting one. What happens when not you, of course, but when staff members hide their use of AI. Right? Like, and I've gotten to the point now where I'm convinced that if I gave two different people a project or something to do in Claude or perplexity or you name it, I can tell that it was I could tell how much help they got from AI largely based on what was the output, and also just context. The lack of context in AI stuff, you get stuff that's very undercooked and very generic. Accountability. Who's responsible when the AI gets it wrong? Right? AI ate my homework. Who's to blame here? Right? What are you gonna do? I work with an amazing CFO, and she uses AI all the time to repair reports and financial things, but she always says, I always double check it. Because you'd think math would be the thing AI would not get wrong, but we could explain why it messes up. You know, check the numbers, but is there accountability? And this last one I want you to think about because I see this for sure. It's what's called the productivity paradox. So what if AI makes us more productive but not more valuable? A big thing that's happening right now is that, a lot of us who use AI feel like I'm more productive. I'm able to do more as an individual. I can get a lot more done. It helps me in a lot of ways. But I'm not so sure it's helping my team move the needle or my organization move the needle. And so that's something we definitely need to keep in AI. We need to keep an eye on is that we're not just becoming more efficient, but are we also becoming more effective? And then certainly, there's other concerns, concerns about the environment, the use of the grid, energy, water, that comes up with a lot of organizations. We could probably have a whole other slide of other potential challenges. My advice to you today is just be aware that these are real things, and things that you need to keep an eye on. Keep an eye on your AI and just be cognizant that different people in your organization may be more worried about this than than maybe you are potentially, and it's something to to keep an eye on for sure. Absolutely. Okay. Let's wrap up here. Karen Lamb, a year from now, you may wish you had started today. So I have thrown a lot at you, and, you know, one of the things that often comes up here is, wow. Overwhelmed, Steve. I got some AI anxiety. I thought I was a fox, but now I'm a I might be a squirrel depending on, you know, how I'm feeling, is what to do next. So here is I'm gonna give you a road map. So you're with products all day long, so we have road maps. I wanna give you a a thirty day AI road map. So how to make you and your organization more AI driven in thirty days than you are today. So let me break this down across a couple of weeks. So let's start with this week. Right? Take the next forty eight hours. Take the next two days. Look through your notes. Think about what we've talked about here. Maybe pick up a copy of AI driven nonprofits. Amazon might be able to get it to you in two days. And just think about all the things that have been brought up here. Right? And then next week, Monday, you're gonna hit the ground running. First, do some diagnostic. Here's what I recommend. In your team meeting, your staff meeting, the next meeting you've got a group of of folks from your organization, Do the spirit animal test, chess. It's not a test. Exercise. And, well, hopefully, that'll give you a sense. Okay. Who on the team that I work with? Where are my foxes? Where are my squirrels? Where are my, rabbits? Right? Do that exercise. Get a sense of where people are today. Week two, do what I call an internal pain point audit. So think about it this way. What are some repetitive tasks that drain our energy, as a team, as individual staff members? What's something we could that could be a quick win? You know, upper left hand box. What's something we could pilot? But tie it back to something you wanna improve. This isn't about, like, AI for the sake of AI. We're trying to drive a better result. We're trying to improve something. So in in week two, think about what is something we could try, what is something we could start doing, what is something we could redesign that could potentially help us? Week three, governance. It's gonna happen, and so what I recommend is do your own red light, yellow light, green light scenario session. Some of you may have real examples of things that you know you have done or thought about doing at your organization. Maybe some of the examples make up your own, but just play the same game. Okay. Let's look at these examples. What do we think? Red light, yellow light, green light? Now the the real trick here is that this is what a lot of governance is gonna come down to. Right? You there's something you wanna pilot. There's something you wanna try. Do the red light, yellow light, green light test. What do people think? And as we saw here today, five different scenarios. Some of you, none of it was unanimous. Although maybe there was one, I think everybody said red, but it wasn't totally unanimous. So the point being, use that to guide, okay, what are policies, boundaries? What do we have to have in place to make it work? Right? And then week four is when we actually put a policy together. And I did this intentionally because if I told you the first thing you need to do is go off and write your one page policy, I would have lost the hundreds of you on the call today right from the get go. Work your way towards that because by week four, getting close to thirty days, you should have a sense of where you are, where your team is, where the organization is, where you wanna try things. Publish a one page, maybe, interim acceptable use policy. If your org already has one, love it. Awesome. Great. What I want you to do is, does it need to be updated? Do you need to cover it? Just because you covered it that one time in a staff meeting doesn't mean everyone's read it. One other thing you could do as you're developing the policy, ask people to play red light, yellow light, green light based on the policies that your org has. But, again, this is a journey, but if you really focus the next thirty days where you are, your team, your org is, and just take it one week at a time, you will be amazed how much more AI driven, you're able to be, in a very short period of time. Okay. I think that's everything. I will stick around a bit, but, love to take questions, answers, anything you've got in the time left. I really try to put as much content in here as possible for you today. So, yeah, I'm gonna take a quick drink, and then if you've got questions, happy to hang up for a bit and answer as many as you've got. Yeah. Okay. Great. Great question. What if your organization's working on a governance policy, but there's a split amongst staff around proper ethical usage? Love it. Right? Again, that's not about AI. That's about how your organization resolves conflicts or makes decisions. And so what I would recommend is, like, have the serious discussion. Right? Where where is that split specifically? And then think about, okay. I know there's a concern in this area. What what what process, guidelines, guardrails would we have to have to move you from being red to yellow or from yellow to green? Talk it out. Right? Like, what is it specifically? Get that feedback. See if that can be potentially incorporated. Ultimately, I would say you get to a point where, leaders in your organization have to make a decision. You rate you weigh all the puts and takes and the inputs and outputs. But I see before you get that far, just ask okay. You got you're I can tell, you know, Bob, Susie, I can tell you're really concerned. You brought this up multiple times. Talk to me. Talk to us about why you're feeling this way, and what do you think we could put in place as a guardrail that could potentially prevent this? Where do you think the risk is? And get those things to surface. See if you can incorporate that into a policy. Ultimately, you know, it may come down to you you've gotta make a decision that we're gonna choose to do something, but that's would always be, my recommendation is a way to think about things. Any other questions? No one's asked me about any of the tech or the tools, so I consider that a victory, at least today because we could do part two of this and it could be all about the tools. But, I find this is a good place to start. Well, Abby, I think we've got all of our questions. I think you've got a poll up, and, obviously, wanna thank everyone for attending. Wanna thank Blackbaud for the opportunity to come here and talk today. And then, obviously, to learn more about the book, you can click the QR code. My contact info is in here somewhere. And, as always, if you're interested about some of the ways, that Blackbaud, can can, your organizations can use Blackbaud and some of the amazing AI tools they've developed. You can always request a demo, provide feedback within here as well. But, again, good luck the next thirty days. I'm gonna check-in the next thirty days and see how everyone's done, but really appreciate everyone's time here today. Thanks so much. Awesome. Thank you so much, Steve. And everyone, thank you for your time. You'll receive the link to the recording as well as the slides, so keep an eye out. Steve, as you can tell in the chat, this was such a great session for everyone, and really appreciate your time. For anyone that was interested, we will follow-up with you directly, to let you know more about our solutions. And thank you, Steve, for the shout out. So, feel free. You can always, message me directly as well, and we hope everyone has a wonderful day. Thank you so much. Thanks, everybody.