What Nonprofits Get Wrong about AI
Ready or not, here come’s AI!
Artificial intelligence is already showing up in human services, whether organizations are ready for it or not.
In this episode of Thinking CAAP, host Beck Moore is joined by Amanda Hoprich and Ryan Peasley from WIPFLI to unpack what AI actually means for human services organizations, including Community Action agencies. Together, they explore common fears, compliance concerns, and misconceptions and clarify why AI should be viewed as an augmentation tool, not a replacement for human work.
The conversation covers practical entry points for beginners, how AI can reduce administrative burden, support grant writing, streamline workflows, and help leaders combat burnout while preserving human-centered service. Ryan also emphasizes the importance of proactive governance and clear organizational guardrails to ensure AI is used responsibly.
This episode is essential listening for nonprofit leaders, administrators, and teams navigating rising demand, limited resources, and the future of mission-driven work.
Click here to learn more about Ryan Peasley and the services WIPFLI provides.
Thinking CAAP Episode 31 Highlights
[0:00] – Introduction to AI in Human Services with Beck Moore, Amanda Hoprich, and Ryan Peasley
Beck Moore welcomes co-host Amanda Hoprich and guest Ryan Peasley to discuss AI’s role in human services.
They introduce WIPFLI's history and Ryan’s leadership in helping federally funded human service organizations adopt technology strategies and AI solutions.
[5:25] – Human Services Reactions to AI and Overcoming Skepticism
The hosts explore common reactions to AI in human services, from excitement to skepticism.
Ryan highlights fears about replacing human interaction and discusses organizations feeling overwhelmed or uncertain about adopting AI due to capacity and knowledge gaps.
[10:00] – Common Misunderstandings and Compliance Concerns About AI
Ryan addresses misconceptions such as AI replacing jobs and compliance worries about data privacy like HIPAA.
He emphasizes AI as an augmentation tool that requires governance and human oversight to reduce administrative burden while maintaining service quality.
[12:59] – AI as a Tool to Reduce Administrative Burden and Improve Efficiency
Beck Moore shares experience integrating technology in large human service organizations, noting AI's potential to streamline processes.
Ryan underscores AI’s role in lowering administrative costs and increasing programmatic impact, advocating for strategic adoption in human services.
[14:50] – The Evolution of Roles Through Technology Integration in Human Services
Beck reflects on how technology adoption shifted job roles rather than eliminated them, creating new positions focused on data compliance and management.
The conversation emphasizes that AI will transform but not replace human jobs, requiring new skills and
[15:06] – Embracing AI for Capacity Building in Human Services
Beck Moore and Ryan Peasley discuss how AI and automation significantly boost capacity in human services by reducing administrative burdens, such as case notes and compliance paperwork.
This enables more direct client interaction and creative process improvements.
[18:20] – Getting Started with AI: Practical Tips for Beginners
Ryan Peasley advises beginners to start using AI tools like ChatGPT for individual productivity tasks such as drafting and summarizing emails.
This saves time and helps overcome writer’s block, making administrative work more manageable.
[22:11] – Leveraging AI for Complex Tasks Like Grant Writing
The conversation highlights how AI can simplify complex tasks such as summarizing lengthy funding opportunity documents and drafting grant proposals.
This allows nonprofits to save time and improve proposal quality through iterative AI-assisted editing.
[25:47] – Enhancing Email Management and Workflow with Microsoft 365 Copilot
Ryan shares how Microsoft 365 Copilot integrates with Outlook to prioritize and draft responses to emails.
This significantly reduces administrative workload and improves efficiency, especially for professionals managing high email volumes during busy periods.
[28:11] – AI as a Modern Tool for Learning and Efficiency in the Workplace
Beck Moore reflects on the shift from traditional learning methods to AI-enabled tools that simplify tasks like Excel budgeting and data analysis.
He emphasizes how AI now accelerates skill acquisition and daily productivity in ways previously unimaginable.
[29:51] – Streamlining Organizational Tasks with Microsoft 365 Copilot
Ryan Peasley describes how Microsoft 365 Copilot automates time-consuming tasks like data compilation, email sorting, and PowerPoint creation.
He highlights its seamless integration and efficiency, especially for board meeting preparations and report summarization in community action organizations.
[33:23] – Enhancing Client Service with AI-Driven Scheduling and Communication
Ryan discusses how AI tools, such as self-scheduling forms and chatbots, reduce callback times and phone tag for small teams.
This improves human-centered service by meeting clients’ availability and language needs.
It enables organizations to provide timely assistance regardless of traditional office hours.
[38:46] – Navigating AI Risk and Establishing Organizational Guardrails
Ryan emphasizes the importance of proactive AI governance policies to manage risks like data privacy and inappropriate use.
He encourages organizations to define acceptable AI tools and uses to prevent incidents, rather than reacting reactively after harm occurs.
[42:48] – The Necessity of AI Policies to Mitigate Organizational Risk
The conversation highlights how many organizations unknowingly have staff using AI.
It stresses the need to implement thoughtful AI policies early to reduce risks.
This helps avoid restrictive knee-jerk reactions following incidents and promotes safer AI adoption in human services.
[43:45] – AI’s Potential to Alleviate Burnout and Amplify Impact in Human Services
Ryan expresses optimism about AI helping human services organizations address staff burnout and high turnover.
This is achieved by automating low-impact tasks, allowing leaders to focus their efforts
[45:19] – Leveraging AI for Transparency and Real-Time Impact in Nonprofits
Ryan Peasley discusses how AI can enhance transparency and accountability for federally funded nonprofits.
This allows organizations to showcase real-time impact to stakeholders.
It helps nonprofits demonstrate their outcomes effectively and move the needle on community missions like closing the poverty gap.
[47:07] – Combating Burnout in Human Services Through AI and Automation
Amanda highlights the emotional and physical toll of human service work.
She explains how AI-driven workflow automations can reduce burnout by handling administrative tasks.
This allows frontline workers to remain present and effective in their service roles.
[49:13] – Addressing Increasing Demand and Resource Constraints with Technology
Ryan reflects on the rising needs for social services amid shifting systems.
He emphasizes that efficient use of AI and technology is essential for nonprofits to serve more people effectively.
This is especially important despite fewer resources and growing challenges.
[50:42] – Encouraging Engagement and Support Through WIPFLI and CAAP Resources
Amanda and Ryan invite listeners to connect with WIPFLI and CAAP for personalized support in technology adoption, especially AI.
They share contact details and highlight upcoming CAAP events focused on AI strategy for nonprofits.
[52:07] – AI as a Tool to Amplify Mission-Driven Service and Upcoming CAAP Summit
Ryan stresses that AI should support, not replace, human service efforts by freeing up time and energy.
Beck promotes the upcoming CAAP summit in Harrisburg focused on AI.
They all encourage organizations to explore AI strategies for enhanced community impact.
Episode 31 Transcript
00:00:00
What's up, everybody? This is Beck Moore, CEO of Community Action Association of Pennsylvania, and your host today of the Thinking CAAP episode. And with me, one of our favorite co-hosts, Pennsylvania's darling. Welcome, Amanda Hoprich. What's up, Amanda?
00:00:13
Oh, not much. I'm very excited about today's episode and co hosting, since I'm kind of straddling both worlds here. So new world meets old world. So, as we said, we'll introduce our guest, but we're going to put him in the hot se. This episode really is around WIPFLI and showcasing some of our work in AI.
00:00:40
If you're new to the Thinking CAAP. Amanda used to be with CAAP. She was my co-worker. Worked together for a long time, and then a little over a year and a few months ago, officially, I think three months. I'm going to say three months.
00:00:54
That's my guess about that. Went over to WIPFLI to work there, which I'll make sure to let Ryan talk a little bit about their company and the great work that they do. Amanda is back and happy to have her. So welcome back, Amanda, and also our guest. Today we're here to really talk about AI, particularly AI and Human Services, which is something that our guest, Ryan Peasley, has spoken about at our conference in different sessions.
00:01:20
I know has traveled the country to also talk about this topic at different times, but we're really at CAAP, trying to think about how to approach this topic. It's something that we're having a summit about in the spring and excited to talk more about that on a future episode. But welcome, Ryan. What's up, man? It's good to see you.
00:01:38
Yeah, thanks. You know, it's been a long time trying to get this. This schedule finally got it on the books. Appreciate it. Not sure I'm ready for the hot seat yet, but certainly here to share a little bit about AI, a little bit about WIPFLI and how we're helping organizations kind of take on that journey.
00:01:55
Absolutely. So can you start by just telling us a little bit about yourself, what you do at WIPLY, And I know you have a quite a large role in all the work that you do and all the different types of work that you cover. So just give our listeners a little bit of a context. Yeah, absolutely. A little bit about WIPFLI.
00:02:11
We started as a CPA for many, many years ago and have had many iterations. We're almost getting to a point where we're 100 years old now at WIPFLI, and for 50 of those years, we've been Helping human service organizations, federally funded human service organizations, whether you're Community Action, Head Start, or other notable federally funded, you know, programs. But outside of that space, we also help healthcare organizations, manufacturing, distribution. The list goes on. Where we have several different other specialties in industries.
00:02:46
And for the last 15 years, I've been helping nonprofit organizations with their various technology strategies. But then also a few years ago, I don't know if you call it promoted or volunteered into a position where I do lead this division at WIPFLI now that specializes in working with federally funded human service organizations, whether that's through audits, financial system implementations, process improvement, work strategy, risk management, HR technology. The list goes on. Right. So here really, the focusing on that technology space.
00:03:31
And I'm sure some conversation around data will go as part of this year today, too. So excited to kind of dig in and see what we can do to help organizations kind of start their AI journey or just continue the journey that they're already on. Yeah, absolutely. And we'll just say, you know, thank you to WIPFLI because you are a big supporter of CAAPs and an important partner for our success, and so just appreciate the partnership that we've established and continue to grow over the years. So we'll just say Shameless Plug here for WIPFLI and all the great work.
00:04:02
That they do as a marketing manager. I love that shameless plug. So.
00:04:11
We're selective in the organizations we choose to really partner with.
00:04:17
We want other organizations to be innovative and thinking through what's next, not how do we check the box or keep doing what we've always done. So, you know, CAAP has certainly, you know, wowed me after I've got introduced to y'all and the things that you're doing for your network and the greater community action network as a whole. So certainly appreciate what you're doing, too. Well, thank you for that. Yeah.
00:04:41
Much to my team's chagrin. I like big ideas, so thank you to them who are always dealing with me. I can attest as a witness to that if.
00:04:54
Yeah, you know, I start out any kind of new hire conversation with. Just want to be really clear. Right. Like, I love strategy and I love ideas. So if you.
00:05:01
I'm not everybody's cup of tea, so if you don't like that kind of leader, then this isn't the right place for you. But we have a great team here. And, Amanda, we still miss you. All right, so let's get into some of our questions, and I think, you know, we'll kind of back up. We really started probably About a year ago to start, you know, there are some discussions with some of our network folks about AI and the importance of.
00:05:25
And one of our board members had said that they had gone to a really just wild conference where the red carpet was rolled out. It was not a WIPFLI conference, to be clear. It was a different financial services company who shall remain nameless. And the production value was like bananas. I mean, the swag that they got, it was wild.
00:05:49
And what part of the main, a lot of the main learning at that conference was around AI. And she and I were chatting a little bit and she said, you know, we spend so much time in the work, we just oftentimes don't have the time to really dig into where we can implement these types of tactics. But I know we have to. And that conversation with Erin has remained on my mind for ever since. And something that we have started to kind of weave in.
00:06:15
But this spring, right, as I mentioned before, we're going to be having this AI summit slash workplace culture, sort of alignment of topics. And I know that within human services there is so much reduction of administrative burden that can come about when we talk about how to embed AI. But let's start with kind of, right, Something that I know is a little bit of a pain point in that when people in human services hear AI in quotes, what reactions are you usually seeing and where do you think those reactions come from? And I, I know I've seen them too, so. Oh, I mean, there's definitely a spectrum of individuals who are super excited and they're trying to force every idea that they have or AI can be put into a situation or solve a problem.
00:07:05
And then there's the skeptics and then a couple of variations of individuals between, let's say the skeptics are still very much talking about, hey, we are helping humans. There needs to be this high touch human component to it. There's, you know, we've got all these emotions that are often involved and vulnerability of, you know, the individuals we serve that coming in, you know, that to force them to use AI just won't work. Right. Well, I think there are some myths in some of that.
00:07:40
There are. That's one of the end of the spectrum for sure. Then I think between those two, there's this group of organizations that are, that are paralyzed because they don't know what to do. Like, is this, you know, a fad? Is this a real strategy?
00:08:00
Yeah, and I don't know what to do. I don't know if I have the internal capacity to take this on, and I don't know who could help me take this on either. So there's so many questions that it's overwhelming for them to answer those questions that they just don't do anything about it. Yeah, well. And I think if you're not comfortable.
00:08:17
So as you know, someone who affectionately gets called pop up in the team from time to time. Thank you, Amanda, for coining that.
00:08:24
It all started with a conversation about bookmarks. We'll tell that story at a later time. It's a good one. But I know that I have a certain amount of bias around certain technologies because I don't have capacity or time to learn the new thing. And so I think about others in leadership space and I think, okay, well, if they're not comfortable with it, then I know the question that they have is how do I manage it?
00:08:53
If I don't, if I'm not comfortable, I don't know the questions to ask. Right. And so it's sort of, there's a big, I think, trust continuum that is created in that space a bit is, you know, one of my takeaways. So go ahead. Were you to say something?
00:09:10
I was just going to add, you know, historically these, these large technology investments have not been, you know, top of mind or even executed as strategies in a lot of human service organizations. So there's, there's just a new mindset of just that decision making and understanding a return on investment from technology is not, you know, it's not the same decision making factors that you might normally be used to in your director's room or your boardroom. So there's, there's different skill levels. You need to kind of equip your board and your directors with, with making these types of decisions too. Yeah, absolutely.
00:09:47
Well, and, and I think to your point, historically, human services world, I think about all the applications I've ever filled out for different things. We, we love paper. CAAP doesn't love paper. We try to reduce as much paper as possible. But in general.
00:10:02
Right. Like greater community, definitely there is a love for paper processes which again. Right. Comes with a different burden of time, energy and cost. So when you think about this in context, then what do you think people misunderstand the most about AI within this space?
00:10:22
I think there's a lot of, there's some, some mindsets where people are worried about it taking their job. You know, there's, as you mentioned, there's, in human services, very paper heavy. And so it's individuals, whether it's paper or now they digitize that paper into a PDF file or whatever it might have been. There's still this manual work that's happening a lot of times and they go, well, if this is automated or if AI can do this faster, where is my job really going to be? Truthfully, AI, there's still a lot of things it doesn't do well or doesn't do right.
00:11:02
And there's just layers of human interaction that need to be involved. And so I say AI will augment our jobs. It certainly won't replace our jobs. We will have to learn new skills to utilize it. Managers of organizations will have to now learn how to manage people and perhaps AI agents together to get the most impact out of their people and their technology or AI investments.
00:11:28
So some of those kind of misconceptions really around the AI will do my job. I think there's, after you get past that, I think there's a lot of people that go, well, there's a lot of compliance concern because if I put in my, my files or if I start having it read the organization data, there's HIPAA concerns or PII or other, you know, data compliance or privacy concerns. And while that is a good, I, I would, you know, high five that individual say kudos for bringing that up. That is a good concern. That is a control we need to take in place as we talk about our journey.
00:12:08
But there are governance aspects that we can put in place with AI, whether it's during the build process of IT or just organizational understanding on what AI is, isn't and how it should or shouldn't be used in our organization. So there's, there's ways to navigate all of this. And I think the key point is we're not taking anything away from what we do as a human service organization. We're enhancing what we can do. Notably, we see a lot of pressure to reduce, I guess our administrative costs to make sure that for every dollar that the maximum 99 plus cents are going to programmatic outcomes or whatever, you know, the right ratio is there.
00:12:59
But AI is one of those levers we'll be able to pull to be able to make that happen. Yeah. So in a former organization that Amanda and I used to work in, that shall remain nameless. Years and years ago, I led the integration of various technologies into a membership based organization that covered 30 counties. And at the time we had 30,000 plus members and then an additional 15,000 volunteers.
00:13:29
And so up until that point, all registrations came through via paper. Three different pieces. Like carbon copy, right? Like top copy. You keep middle copy goes to another volunteer group and then the organization keeps the bottom copy.
00:13:45
And so we had literally hundreds of filing cabinets across 12 different offices that had every paper copy of everything that had ever been filled out in the history of the organization. And so suddenly then we're asking people to go online and fill out registration. And the fear that I saw and then, you know, of course what was starting, started happening in the very beginning is that paper copies would get lost to get inputted into a database. And so people were then taking copies of forms. And then finally we got to a point where like the issues were, were solved, but people were still hanging onto the copies.
00:14:21
So I've lived through the integration of really large technology systems like Salesforce organization that historically did not use technology. And I've lived to tell the tale, but what I saw in that experience, and my point of sharing that story, is that we ended up hiring more people long term in different CAAPacities. Right. So the same people that were in charge of like just entering forms, sure. Those positions evolved, changed.
00:14:50
They were around data compliance. Right. But we spent then more money on the people then that were actually in the field supporting the volunteers and doing the programming side of the work. And that's really what I like in this too as well. In a lot of ways.
00:15:06
I just, I'm going to pause to back up on one thing. First of all, can we just recognize Ryan's giving virtual high fives for data compliance as one of his main roles is cybersecurity, one of his day jobs, Which I love just saying, very important. But I think the other piece, and to your point, Beck of like, you know, we're talking a lot about compliance, a lot about some of those things that the conversation also naturally sort of goes to like the big buckets of generative and automation and what. And implementing those things looks very, very different, but those are the most like recognizable implementations, AI and that especially when we're talking automation. To your point, Beck, it is such a capacity builder, I will say, like, you know, my current position at wifli, even in the, you know, not having a very long tenure, my capacity has been a hundredfold gain by the way, that we've been able to embrace utilizing AI and then, you know, creatively utilizing it for our processes and automation.
00:16:23
And I think that's the biggest impact. When I think of, you know, I'm the big. Beck likes to dream big. Ryan's in compliance. I'm like the starry dreamer in the middle here.
00:16:32
Bringing it all together is, I think, about those, you know, especially in human services, starting my career in direct client service way, way back when is oftentimes I, you know, I only got 20 hours a week in actual work with families because 20 hours a week were spent on filling out all case notes, filling out all forms, filling out all of these pieces with compliance where if you could actually interact with the family and have a piece of technology that is, you're just asking questions and being able to pull that in and it's going to give you those notes in a summary that then obviously need a touch. The capacity gain is the things that put stars in my eyes in a lot of these things and to your point gives more higher ability, more opportunity. Well and I also think about the way in which we manage in human services and it's hard to know how effective somebody is unless you have a story. It's hard to weed through all of the data. And so I think about from a management perspective as something that I, you know, coach and present on all the time, like to know how fast somebody is following up with somebody and whether or not they have capacity.
00:17:48
And seeing that in data analytics, to me just changes the game in terms of how we can manage in this space as well. And very few organizations today are doing that in this space because the systems, it's not about not tracking outcomes. We track more outcomes than you can shake a stick at. It's more about the pace of follow up, how effective somebody is, all of those things. So Ryan, for somebody who wants to start using AI tomorrow and let's assume that they're at the very beginning of their learning journey here, right?
00:18:20
Maybe they've not even heard the term chatgpt. Maybe they have, I don't know, see enough commercials about it. And ChatGPT, if you're listening, feel free to sponsor the thinking CAAP. But without disrupting anything, what's the simplest way to begin? Where, what would you advise?
00:18:38
I think where a lot of people start and have early success was where I would tell everybody else to kind of start looking at it too is in that individual productivity area. Have it start drafting emails or summarizing. Summarizing the long emails because one thing people understand at WIPFLI about me is if you write a long email I am not reading, heck, I'm probably not reading your email anyways. But long one, I'm certainly not reading. It two paragraphs, that's all I've got in me.
00:19:09
Yeah, exactly. But so summarizing those longer emails, when somebody writes you that book around, whatever it is they have urged to tell you about that day or when it's drafting up a communication for a particular stakeholder, like, hey, I want to write a donation letter or I'm going to be meeting with this local elected official. Can you help me create a few talking points? And you start by doing that with, with a tool like ChatGPT or Microsoft Copilot, Google Gemini Claude is another one that I see a lot of organization use. Perplexity is out there all over my Facebook and Instagram ads all the time too.
00:19:55
All of them are good. Pick one. Doesn't really matter which one you pick. You'll. You'll start exploring others once you start seeing what you know you like and don't like about the one that you picked ultimately.
00:20:06
And then that will kind of start getting your creative juices flowing on how you can use it more and you'll start looking at, oh, I need to provide a report about these five programs to my board member. Here's the, here's the five talking points I have or here's even some of the Excel spreadsheets I have. Can you help me do this? And it will create that narrative, you know, to talk to the board member or community stake, some community stakeholder group or whatever that might be. And you'll start to go, well, it would take me two to three hours the course of the day to do all these extra things that I had AI draft me.
00:20:44
And so you start, you want to start looking at. You don't need to track it any particular way, just I do more of that. All right. Yeah, that was about two to three hours today. I saved using this.
00:20:54
You start adding that up over time. It's insane at what else you were able to get done during those two to three hours of. Not that it wasn't high value work because you're talking to a board member or a local elected official or whatever it might be. But now you're able to actually get more focus time on solving that problem than just sorting through all your, through all your emails. Yeah, you're able to, you're able to focus on the things that matter for your job.
00:21:23
You're not doing the administrative because writing an email, whether you like to say it or not, it's administrative in nature. Right. Solves the blank page syndrome. It solves the, you know, I do like to imagine the days where I was sitting in an office drinking my coffee. Like I have to write female to board members, you know, And I, I know for myself one of the biggest things is I, I also speak To AI like that, I will say I'm stumped.
00:21:51
Here are the things I need to do. How? Yeah, I mean, the more prescriptive you are, right. The better the solution is that you get back. So if you need an email that is very specific in tone, that gets this message across, but not too complex, make sure it's at this reading level.
00:22:11
Make sure you weave in this story or this point, by the way, I call this person this. The more prescriptive you are, the better your solution spits out, which I love. And you can keep telling it to rewrite it and add in this point and remove that point. One of the things that I've heard somebody say it might have been Georgia, our chief strategy officer. I can't remember who it was, but, you know, when you get a notice of funding opportunity and it's a thousand pages, drop it in ChatGPT and say, please summarize these particular points so that I can best determine whether or not our organization is able to apply genius.
00:22:48
Right. I mean, it's a lot. Just the comprehension that it takes to do some of this work to simplify that Holy moly. Translates legally is.
00:23:04
Yeah, because you start reading those no fos and you're like, what did it say now? Like, all right, you go back and you read that paragraph three times. You're like, I don't. Is that me? And you're 75 pages in at this point.
00:23:16
Right.
00:23:19
And by the way, the turnaround on those then funding opportunities, sometimes they're right now are like, you know, it could be two months, it could be a week. It just, it depends. So if you can save the time. And then you have to get various directors involved, you need your development team or who assumes a responsibility of some of that. What development team is if you don't have that in your organization.
00:23:44
And we know with busy schedules, especially with nonprofit organizations over the last year, all the other battles they've been fighting, get that group together to focus on that. You know, that proposal that they're going to write is not easy. So having a quick, hey, let me understand the key points of this and then maybe help you draft, you know, that proposal. Not that you would rely on it, copy, paste, send. But I always like to say it's a drafting tool.
00:24:11
It helps, as Amanda said, get rid of that blank page. And now you're able to start going in and adding to it or saying, ah, that's not exactly accurate. Let me put my actual organization's bio in here. That's more up to date. Whatever it might be, but you're saving just hours upon hours upon hours just with something as simple as NoFos and grant writing.
00:24:33
Yeah. Recently we put a proposal in for a project and we had the scoring criteria. So I said here's the scoring criteria, here's the proposal. Score this based on these things, assuming no one knows anything about your organization. And so it gave me an editing opportunity to go back and refine what I had proposed.
00:24:52
And it had already gone through multiple iterations at that point of people and rewrites of different things, but that was incredibly helpful and not something I had had done before a lot of times because simply there is no scoring criteria that is shared. Any other tactical examples of other tools? You know, let's think about maybe some of the more, you know, getting to the folks who maybe, you know, they've, they feel like they've maximized ChatGPT and some of those, those tools and we start to creep into kind of the middle, more comfortable folks. Any other tactical examples that you care to share? One of my favorite ones, again, we talk about emails and you know, I do get to the benefit of traveling and going to various conferences, whether it's the CAAP conference, I'll be going to a community action conference next week where I'll be speaking on AI here too.
00:25:47
And I just can't keep up with my, my emails when I'm out at those counters. I come back and there's several hundred more emails to go through and parse through and try to prioritize and respond. And so one of my favorite Tools is Microsoft 365 Copilot, and it integrates with Outlook so when I'm in there, I can prompt it. Just read all my unread emails and prioritize them for me. And then even are there any I need to follow up with, you know, the next 24 hours or whatever that time frame is.
00:26:21
And it will spit up all of this, you know, like, Here are the 30 emails you got to get to today. I'm like, oh great. And then I can even prompt it though. Can you draft a response to these 30 emails? I might not like all 30 drafts of that, but hey, I started the draft and can confirm whether this is good or not and click send later.
00:26:43
But that helps me tremendously from, you know, the administrative level that's kind of like that next level up. Then just, you know, a prompt and getting an output and copy and pasting it somewhere, which kind of think a lot of people do with, you know, chat GPT at first the whole Microsoft 365 copilot integrates with Word and Excel and PowerPoint and several other apps inside of Microsoft 365 ecosystem. So if you're an organization that uses Microsoft 365 for email or file storage or all the other things that, you know, it can do, Microsoft 365 copilot is probably a quick add on, you know, for you. There are some free versions where it's just Microsoft Copilot which is chat based, much like chat GPT. Microsoft 365 copilot slightly different variation in the name there.
00:27:37
But leave it up to Microsoft to make licensing confusing. That's what they do. But that at, you know, the, you know, 20, $30 a month that it costs you, again we are talking about saving two to three hours. I mean people will pay that, you know, to, to get that functionality along. So whether it's drafting PowerPoint slides, creating word documents, trying to figure out how to create a pivot table in Microsoft Excel, like all that stuff can, you know, AI helps with in that Microsoft 365 copilot space.
00:28:11
That's pretty slick. Yeah. You know, I think so much about was just having a conversation with somebody about this the other day when I graduated college. I had never used Excel. There was no reason for me to ever use Excel in college.
00:28:23
This wasn't English majors. Right. I mean I was humanities. English Humanities, whatever my degree is in. And it was my very first job and I had to do all this budgeting work and so I had to go to a community college and take an Excel class because it just was a different time.
00:28:43
Right. Like Google just wasn't a thing. Yeah. And maybe it wasn't. I just.
00:28:48
Now is pop up.
00:28:51
Right. I know. Although I was on a call this morning with a guy and he's like, you certainly can't be older than I am. And I was significantly older. So thank you very much.
00:28:59
You know, Sox soon but I think about how much time I spent trying to figure out how to do stuff. Yeah. In. In a world now where it's just like, right. I just, I can Google it or I can just tell my computer play, please go do this for me.
00:29:19
And not to give another. Right. Not to give another email example but like the other, the other. Literally yesterday I, you know, we're gathering, we're looking, you know, ahead at the WSLI conference in July and we're going through some selection processes for some of our speakers and I had had suggestions sent to me all year and I was Able to say, you know, look for anything that has the keywords of XYZ for stronger to serve. Plop that into an Excel.
00:29:51
Please give me an example of work. Give me all of these things. And then my output is this beautiful Excel with all the things that would have taken me a day.
00:30:04
Putting all of those things into a folder. Check out this folder in my email. And then put all of this into a digestible PowerPoint. I do that constantly. Hey, everyone, send me your ideas for xyz.
00:30:17
I just create a rule, put it in a folder, and then when I'm ready or when I need that output. Microsoft 365 copilot it's just seamless in being able to do that. That is probably the AI tool I use. Absolutely. Above all else.
00:30:36
I got examples of more specific to the community action space. I was just speaking to, actually two CEOs in the last day now. And one was talking about how they used to spend all this time drafting, you know, board content for their board meetings. And they'll draft the narrative to particular things for the board. And the board does not like.
00:31:01
They're like, hey, this is efficient. Actually, you're putting it in the bullet points for us. We'd rather read bullet points than paragraphs. And so she's been, you know, excited about, you know, doing that, and she's like, it saves me three, four hours of month just doing that, let alone all the other things that I'm, you know, I'm using with it. We're piloting.
00:31:21
Sorry, I'm going to cut you off. We're piloting right now with software to do our own board work. And one of the, you know, some of the tools that are in there, it's, you know, using AI to summarize a P and L. You know, here are the things that you need to pay attention to. And now suddenly, you know, the finance notes are far more simplistic and take far less time. There's also mechanics in there to take votes ahead of time so things can move to a consent agenda without ever, anybody ever having to talk.
00:31:50
And there's lots of examples of that type of software out there. But that I think is particularly those agencies who are meeting on a monthly basis. I say this all the time, that in theory then, or in practice, not even in theory. In reality, they are spending literally almost every single waking day preparing for a board meeting, collecting all this information. And so that's all they end up doing.
00:32:15
And that executive director then is exhausted, as is their board, by the way, which probably decreases engagement in a lot of instances because then all they're doing is sitting in board meetings and reviewing committee notes and there's not been enough time to do the work, which is a whole other conversation and soapbox that I can get on. But anyway, I think those types of streamlining is, that's, you know, again, I think the, to me, kind of some of the bare bones that I'm hopeful that people will start to take advantage of.
00:32:47
Were you going to share another example, Ryan? Yeah, I've seen a lot of just, you know, from the HR space people drafting and you know, manuals for, you know, HR manuals or employee handbooks, policies, procedures, types of things. Obviously you still want to get those reviewed, make sure they abide by, you know, local laws and all that stuff too. Again, AI as a drafting tool, not as the end all be all there. And then in a lot of program directors, there's tons of news going on out there about various programs and funding and perhaps different compliance things that might change.
00:33:23
And so I've seen some program directors actually go out there, look at what's, you know, whatever the new federal guidance might be on a particular program and they just say what changed from this version to this version? So they're only looking at the changes versus having to consume, you know, the entire document. So some real world examples, you know, perhaps for human service organizations that they might be able to pick up on. I know my wife's organization, she runs a housing organization. She was on the podcast with Amanda as a matter of fact a few episodes ago, she runs an organization, she runs an organization called Echoes and they put in place place they found that she sort of was observing their intake process and trying to follow up on calls.
00:34:03
And what she was seeing was that to get back to somebody who was just scheduling an appointment was taking a significant amount of time. And they're a small team, they have, you know, maybe five people but their specific client facing team, there's only two or three of them on a good day. And so they just put in a simple form schedule, right for people to be able to self schedule based off of availability. And now that's cut down on the callback time. And the team isn't playing phone tag for hours on end on any given day they're able to actually come into work and already have their appointment scheduled.
00:34:39
They might have to follow up to confirm, but there's also, you know, systems in place to be able to do that as well. And so that for an organization that is, you know, just Over a million dollars in a small team that has reduced pretty dramatically something as simple as the follow up. Because the other thing is, and I think this is important for human services folks to think about, is that the clients that the participants of the folks that you're serving, they're working other jobs, they're working maybe 1, 2, 3 jobs. Their availability may not match up with your organizational availability. And so right away, while I appreciate you kind of going back to one of your original points, Ryan, that people may feel like you're losing this human approach or human centered approach, I would argue that it also increases your human centered approach because now you're creating availability, opportunity following.
00:35:29
Right. Like after you're closed, when people may, that may be the only time that they have to set up the meetings. Exactly. You're constantly trying to find out new ways to meet people where they're at. Right.
00:35:41
And so somebody has a need at 3am Maybe the chatbot is their best available option to navigate some sort of crisis that they're in. Sign up for a program so you know when you are open that they can bring their child into the Head Start center or whatever they need assistance with. And we really are seeing a movement towards all of the things that we consume as an individual in our lives. You know, we want to click a button and it shows up at a doorstep in two days. Right.
00:36:16
That's how the world is starting to work or expect not even starting. That's really the expectation with a lot of things too. So when, when I have a need for a rental assistance check or to help pay my utility bills or whatever you know, that situation might be, you didn't help me within 48 hours, maybe even a week. And now you've created more burden on me as a perception. Right.
00:36:42
And there's the various poverty simulation trainings you can go through. I think many of our organizations or even listeners here have probably gone or seen or heard through those. It's like you're just adding layers onto that when you can't serve somebody. So now you've added to their problem versus actually helping them navigate the problem. And so AI, there's a human component to what you're doing for sure.
00:37:07
But you know, AI can help you meet those individuals where they're at more consistently. And it just can. It eliminates barriers. Right? It eliminates barriers that we can eliminate just because of capacity.
00:37:21
And you know, while there is some argument on like AI hallucinations and things and biases it can build. It also sort of eliminates biases Right. Like of the opportunity that you're just, you know, especially for a data input or a scheduler or what that is, you know, whether that is someone who. It was a traditional call intake. And I'm so hard of hearing right now, my volume's up so bad.
00:37:46
But like, so if it's. You get me on the phone and I can't hear what you're saying to repeat over and over what that looks like. And all of our phones and technology now have like preemptive text and being able to fill out simple chatbot scheduling forms. When it's asking your name, it's going to preemptively put my name and reduce it. Just completion continues to eliminate barriers.
00:38:11
100% different reading levels, whatever, you know, you need to get to. To meet that. That population that you serve. It certainly helps. I work with an organization that serves.
00:38:27
They've got it documented. They serve individuals that speak 17 different languages. They have two that they've used to, you know, English and Spanish that they've had most of their forms and now they've got forms in every single language. Yeah. That's remarkable.
00:38:46
So exciting. Since you touched on this a little bit, let's talk a little bit about some kind of the risk issue and sort of the not. Not compliance, but more so I would say sort of navigating the risk and being thoughtful about the risk. And so when you think about the guardrails that really matter when introducing AI into an organization who maybe has not really implemented it yet, now, my guess is there's probably somebody in your organization that's using it. You just don't know.
00:39:12
And so what are the guardrails that really matter that organizational leadership needs to start to think about?
00:39:20
Yeah. First, first thing is I guarantee you that people are using AI in your organization and you just, you might not know about it. So you want to start establishing what AI is and isn't and what it can be used for in your organization. Just like you, you probably have some of that already, or I hope you have some of that in some sort of acceptable use policy when it comes to technology in your organization. Hey, you have the Internet as a tool to do your job, but we don't want you on there posting on social media and shopping the whole day.
00:39:57
You're supposed to use it for productivity. Right. And so the same thing from, from an AI perspective, it's. AI is just another tool. It's just so transformative that we're giving it all this press and all this hype, which rightfully so but we want to treat it just like we have treated a lot of the other technologies in our organization.
00:40:17
And if you don't have an acceptable use policy, you don't have an AI governance policy. Guess what? You can ask your favorite AI tool to help you draft one, or you can reach out to me and I can help tell you what you should or shouldn't have. And a lot of those policies there too. But that's one of the things that I think a lot of organizations don't want to start with.
00:40:41
You know, the conversation of AI governance isn't as handsome as like, hey, look at all the cool things that we can do with it. But that's, that is key to at least establishing how AI can and cannot be used in your organization. And then some organizations even take it a little bit further. Here are the AI tools that are acceptable to be used for these particular things. And it, and it just kind of establishes that, that governance policy, that, that really is going to help, you know, make sure that people aren't going outside the box, causing additional risks by putting potential HIPAA information on a public, you know, AI model of some sort that then sits there and that AI model learns from that information.
00:41:26
It's not going to, it's not going to share that I had surgery last week on, you know, a certain, you know, but it's going to at least share, it's going to learn from all of that data and then use that as part of its response perhaps down the road. You don't want to be in that. There's data privacy concerns that your organization, especially some of the states, have stricter guidelines around data privacy that you don't want to get involved in as a nonprofit. You want to be in the news for the right things, not that right. Right.
00:41:55
You have also. Yeah, the good impact, not the unfortunate part. So that's one of the key risks I see is just not putting that time up front on that piece. Yeah, yeah, no, and I think what having worked in human services and operations for a long time and working in large networked organizations, I think what often has happened is that the risk and the policy comes following an incident and then so it sort of becomes a knee jerk reaction which becomes more restrictive. And so I think, you know, best advice here is again, you know, you know, you have, whether you want to admit it or not, you have somebody in your team that's using some type of AI in some way.
00:42:48
And so setting the policy now that reduces risk, is thoughtful about risk. And isn't the knee Jerk. Based on something happening and not maybe having what feels like capacity to work through the problem in the policies. I think really important for sure.
00:43:09
So as we get to a close here, Amanda, any questions you want to ask? I have probably two more in the, in the tank here, but if you have any in particular that are burning, I want to make sure to give you the chance to ask. Oh my gosh. No circle background. I'll think of a couple.
00:43:26
So Ryan, what when you think about the potential in human services to utilize AI when it's done thoughtfully, ethically, right. Mitigating risk concerns, what excites you the most about it?
00:43:45
Anything in particular? Like. Yeah, I mean, yeah, every, every leader I've talked to in the last year, every conference I've gone to in the last year, all of them have mentioned burnout. I've got staff turnover. They're just, they can't continue to keep up at this pace.
00:44:09
They can't. All of these, you know, various outside influences adding pressure to my organization that are making it difficult to continue serving the community. I can't, you know, I've got, just had talked to another director this morning and I have 90 some positions that are open for my organization. How am I supposed to fill those?
00:44:31
And I think AI, it's not going to be the end all to like, I'm not just going to say, hey, you build this AI tool tool and those 99 positions that you have open are go filled because AI is doing it. But I think it helps us reframe our work to make sure that we are putting in, you know, the, the highest impact possible to deliver those outcomes. So then, you know, you don't leave work at the end of the day thinking like, well I didn't get anything done, I didn't help anybody and all I got was more paperwork on top of this. And so I think AI can certainly help reduce, I don't mean it by replace jobs, but reduce the number of individuals to really create a great impact, you know, in a community. I think that's going to be a huge thing there too.
00:45:19
And then whether we, we really like it or not, we've seen some changes in how, you know, the, how nonprofits, specifically federally funded non profits are viewed in the last year and even more notably in recent news with around, you know, transparency and accountability with funding. And I think if we can layer in AI to really showcase, you know, real time impact that we're providing to whatever stakeholder it is, whether it's our local community, the funding source itself, you Know, the states themselves, you know, and a lot of them help develop these state plans on how this funding is going to go through. If we can showcase that in a lot more real time or closer to real time impacts, we're gonna, we're gonna look amazing. Like, whoa, look at the, the things that we're able to do with a hundred bucks and what it goes for, you know, into serving however many families and look at these outcomes they provide. And that's really going to start moving the needle, I think with how we really start making impact in our community and closing this poverty gap or whatever your mission might be for your organization.
00:46:33
This will start to happen maybe not automatically for us, but it'll start giving us a little bit more of an easy button to push once we develop that. So those are the things that, you know, super excite me is kind of solving that burnout problem and then showcasing what our organizations really do and not keeping it the best kept secret anymore or being in the news for the wrong reason, as I mentioned before. Yeah, for sure. Amanda, were you going to say something? I mean, you know how easy it is for me to get excited.
00:47:07
But I think what really like, you know, I found a sparkle in when you said burnout, right. Because I think the other thing is that like just that exhaustive nature of human service work, right. It takes a lot of empathy. You shoulder so much emotion. It's why I knew, you know, one year into direct service it was burnout like and you know, it is if it can organize your notes better, give you the to do list or you know, workflow automations that are going to trigger, if I'm a Head Start teacher, just by marking who is present and who is absent, it's going to keep me in compliance of so and so missed three days and it's going to automatically trigger a workflow to their case manager.
00:47:56
Like the little things that take you away from the work in that moment that are the things that stress and burn you out. Like, oh, I got to remember to call so and so and note this and I need to check, I need to check these five different things, especially in settings where it's like Head Start or Direct Service in any program where you just want to be present and your effectiveness to be present is what someone needs at that moment. It can start eliminating and shedding those pieces so that burnout doesn't feel the same and everyone needs an emotional break, but it starts to take those stressors away in a way that I think it just to your point, Ryan, like It's incorporated right into the job now. And those kind of investments in technology and seeing those use cases in AI have been really, really rewarding in my short time at Peasley. I'll say.
00:48:56
Yeah, Yeah. I mean, I think 1,000% over. It's the amount of emotional, physical burnout. The number of people who I know have had major health issues.
00:49:13
Heck, the number of people I know who have passed away in the last year and a half is. It just feels different. And I don't know if it's just because I'm getting older that I'm looking at it differently, but I also think that there is just something different in the nature of the work in the last few years that just is all adding to this. And I think anything that helps to reduce that burden is incredibly important, for sure. And I, again, I say this, I've said this a couple times to people, is that at the end of the day, the number of people who need help and support is going to continue to increase, you know, unfortunately, because there's just shifting systems right now.
00:49:59
And so we're going to see outcomes that we don't even know based off a change of one thing into the next thing. And this is not a political statement, to be clear. It is just simply, I think, a point of fact that we will see a difference in the types of services we need, the types of things that people are going to need, and we're going to see more people that need help. And so our ability to serve those people is reliant on resources, which there are fewer of them and us to run more effectively and efficiently. And I truly feel like the only way that that's going to happen is if we think about how to use these tools in the best way possible.
00:50:32
And again, to your point, Ryan. Right. Make sure that we reduce risk so that we're not seen in the news for the unfortunate things, but instead the good things.
00:50:42
All right, last question, because we're coming up on time here. Any shameless plugs that either one of you wants to make about WIPFLI, personal or otherwise? I mean, hot seat, Ryan, favorite co worker. No, just kidding.
00:50:59
But really, I mean, I. I am. You know, I won't give my general marketing spiel since this is the CAAP audience. And as Pennsylvania's former darling, I think if you have, I'm going to make Ryan blush a little, too. If you have a question, I would say reach out to Ryan and the team. Honestly, it's been so fortunate to work with such incredibly thoughtful folks at wifli.
00:51:26
And that truly no question is a stupid question. If it's down to the tiny minutia, it's probably or if it's something that our team hasn't thought about, they'll work it through because everyone's application, especially around technology, has to be so nuanced. I would just. If you just have a question, it's easy to find any of us, but I can at least route you to Riot to get the answer. If you can't find them, feel free to email us@infotheCAAP.org, and we're happy to connect you to the powers that be at WIPFLI, Ryan, Amanda and otherwise.
00:52:07
Ryan, any shameless plugs? I'll just say, as much as we focus, you know, on AI here today, it still gets back to your mission and it's really serving the individuals and in your community. And I understand AI can't serve, you know, the people in your community. That's you. That's you and your the rest of your organization that can do that.
00:52:29
But I really think if your organization is willing to build an AI strategy, it's going to give you more time, more energy to actually go and serve the community the way that you desire to do so. And I think that's worth exploring for any organization. And if any organization does want to explore that, you know, go ahead, reach out. We're willing to help help you do so. And I'll just put in a final shameless plug for CAAP, which is that we have our Like I said before, we have our summit coming up in the spring.
00:53:01
It will be a large focus on AI in particular that'll be hosted this year in Harrisburg at the Best Western premiere. We'll be sharing dates, times, speakers, all of those things soon, so for our listeners, stay tuned. We'll tell you more as soon as we have the more figured out and appreciate everybody listening. Ryan and Amanda, thank you for joining. Amanda, thanks for co hosting today.
00:53:23
Always good to see you both. And thanks for spending a little time putting on your thinking CAAPs. Yes, thank you very much. Catch everybody. Thanks for having.
00:53:33
Thank you.
If you have any Community Action questions you’d like Beck or one of our local experts to answer in a future episode, please email your questions about Community Action to info@thecaap.org — subject line: Thinking CAAP.