Community IT Innovators Nonprofit Technology Topics

Nonprofit AI: Environment, Bubble, Survey

• Community IT Innovators • Season 7 • Episode 34

Share episode

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

0:00 | 27:42

Carolyn Woodard covers three topics shaping the nonprofit AI conversation right now: the real environmental costs of AI infrastructure, whether the AI market is actually a bubble, and a new survey that gives nonprofits a direct voice in shaping what comes next. This episode helps you move from anxious or skeptical observer to informed, active participant.

This episode covers:

  • AI data centers consume enormous energy and water, but the environmental picture is more nuanced than early viral statistics suggested. Cooling accounts for roughly 40% of data center energy use, and about 80% of water consumption comes from electricity generation, making this a grid problem as much as a local one.
  • The AI market is consolidating faster than previous technology cycles. For nonprofits that have been waiting for the dust to settle, the more useful question now is: what is the cost of continuing to wait?
  • 82% of nonprofits are already using AI for internal operations, per Fast Forward's 2025 report, but nonprofits using free, off-the-shelf tools are less likely to have policies or risk controls in place.
  • The Fast Forward 2026 AI for Humanity survey interest form is open now. If you are the person at your organization who knows the most about how AI is being used, your participation shapes how funders and policymakers understand what nonprofits actually need.
  • Practical next steps suggestions: develop a policy, identify low-stakes pilots, check in on AI regularly at all-staff meetings, and ask your vendors about their clean energy and water commitments.

Resources Mentioned:

Nonprofit AI podcast new every Tuesday.

_______________________________
Start a conversation :)

Thanks for listening. 


Carolyn Woodard

Hello and welcome to the Community IT Innovators Midweek Nonprofit AI podcast. My name is Carolyn Woodard. I am here as always with the caveat that I am not an AI expert. I am curious about AI. I use AI a lot in my own work, and I am seeing a lot of nonprofits talking about what they're doing with AI, going to conferences where it's all about AI. And I'm trying to keep an eye on the philanthropy angle of what funders are promoting nonprofits around AI use.

Carolyn Woodard

So with that in mind, this podcast tries to share an interesting recent news article or item with you and how that relates to nonprofits and AI, and any resources that we can share with you and answer any questions, QA that comes in. So you can always get in touch with us via our website, communityit.com, on Reddit slash R slash nonprofit IT management, or you know, respond to the podcast as well. So

Carolyn Woodard

Today I have a couple of resources and news stories to share with you. I saw this interesting article from the Atlantic that was titled "About That AI Bubble." And so I wanted to talk about that a little bit. I wanted to review some of the environmental concerns and uh to also talk about a new research survey that is going out to nonprofits that you might want to take part in. So uh

Carolyn Woodard

To begin, uh let's talk about uh data centers and energy. So this was a recent uh NBC News uh meteorologist segment on what data centers are and why they use so much energy and why they create a lot of heat. And I just thought it was really interesting. So I'll include a couple of resources uh besides that video. It's you know very local to me, so not uh maybe useful to everyone, but I just thought it was a very interesting mainstream explanation of kind of the basic environmental issues around it. So uh

Carolyn Woodard

If you think about it, uh my son is home from college, and when he is running games on his laptop, his whole room gets hot. If you've ever sat with your laptop on your lap and it starts to get hot when you're doing something, watching something, um, the computing generates heat the way that it works. Um, so cooling, uh

Carolyn Woodard

If you imagine a data center with millions of laptops and it all generating heat because they're doing even more than your laptop is doing, they're doing just millions of computations per second. Uh, that's why it's creating all of that heat.

Carolyn Woodard

It's really kind of a return to square one. When Community IT, when I started at Community IT, the thing that we did the most was wiring offices. And if you remember back in the day, there was a server room in the office, and the local server rooms would have various degrees of cooling ability. Your server might be old or out of date or brand new, so it would have different energy and heating and cooling requirements.

Carolyn Woodard

And one of the reasons that going to the cloud made a lot of sense for nonprofits is then you don't have this distributed servers in every office building in the weird closet with all the wires. You could work in the cloud, means you can work from anywhere, and those server rooms would be located somewhere together.

Carolyn Woodard

So the cloud, running the cloud itself, takes a lot of servers and a lot of data centers. Um, but when it was just the cloud and YouTube and you know all the other things that we do online, it was uh used a lot of energy. But then when AI was introduced, that just exploded the amount of energy. They used the data centers that were at capacity, so new data centers had to be made.

Carolyn Woodard

I read somewhere that they put the data centers where the people are. So some I was reading this article about why can't we just put data centers in caves? Like, well, you can put them in caves for the cooling, the automatic cooling of putting them underground if there are caves near population centers where a lot of people are using AI.

Carolyn Woodard

One of the reasons that Northern Virginia, where I live, has so many data centers, the most in the world right now. I've been it's been explained to me because the Atlantic, transatlantic cables come through here. There's Washington, D.C. is right here, so there's a lot of AI usage. Uh, at least the companies know that there will be, for the foreseeable future, an AI usage issue. So they're building data centers at a very fast clip. Um so

Carolyn Woodard

Cooling requirements account for about 40% of the data centers' energy use. Very early on, there was this kind of viral description of if you ask one prompt, you've just spilled one bottle of water out on the ground, you know, uselessly, trying to draw that analogy. And that really went viral. And I think a lot of nonprofits have really absorbed that statistic, even though that statistics itself pretty quickly became pretty controversial.

Carolyn Woodard

And that statistic is from several years ago, so a previous generation of Chat GPT, basically. And of course, exponentially the tools are getting more sophisticated and can do more, they're also getting more efficient. So it doesn't, it's not as hard for Claude or Chat GPT today to return an answer as it was three years ago. So that one bottle per prompt statistic is out of date. Uh, there's lots of,

Carolyn Woodard

I'm not going to go into all of the controversy over what is the actual number, because there's a lot of players that are trying to minimize it. There's other players that are trying to kind of maximize it, as we should all be so worried about this. So you can do your own research if it is something that's very important to you. If you're an environmental nonprofit, you may already have done a lot of this research. You can share it with us. Uh,

Carolyn Woodard

One of the key findings is that uh, as I said, 40% of the data centers cooling, energy is in cooling. About 80% of the water use at AI data centers comes from the electricity generation. So the data center cooling itself is about 20% of the water usage.

Carolyn Woodard

So it's a grid problem in addition to being a local water use problem. Uh it's if you're at an environmental uh nonprofit, then the AI data center growth may be a new lever to kind of use to push for clean energy for the grid in general.

Carolyn Woodard

This is one of those issues that can be looked at in the macro, but it's usually a very local issue. And one that the big tech companies that are heavily invested in AI and data centers are also really, really aware of in that uh the way they get their energy, if that's unsustainable and very expensive, they're gonna find ways to minimize those costs.

Carolyn Woodard

And if they are trying to build data centers in drought-prone regions where water is very expensive, then they're gonna move to the most efficient, least expensive way to use water. So closed systems, closed loops, uh, not having a lot of evaporation if you're in the desert. You're gonna want to build those data centers differently than you would buy them if you're in the Great Lakes region, for example. Um, so uh

Carolyn Woodard

I said in a previous podcast that this is something definitely to be aware of and cognizant of. And if the environment is one of your most important values, then you're gonna take this very seriously. On the other hand, it makes sense to do that as an informed consumer. So really understanding all of the energy use that we have. You know, do you fly somewhere for a conference for your organization? Do you um, you know, serve hamburgers at your gala? Because the you know, beef production is one of the worst usage of water and energy that we have. So kind of taking it all into account.

Carolyn Woodard

And like I said, maybe this is one of those fulcrum that can be used to move to more uh renewable energy that would uh affect the whole grid that those data centers are on. And the data centers are such huge players that getting them on board for cheaper energy for themselves might be something that could work in our favor. Uh definitely, if you are concerned about the environment, uh contacting your local representatives, uh, you know, county, town, uh, congressional uh representatives, statewide representatives. Um

Carolyn Woodard

I know in Northern Virginia we have several congressmen who are and congresswomen who are really, you know, this is a huge issue for them in their district, so they are really trying to lead the charge, both at the state, but also at the federal level to do more regulation. Uh so that is a place that you can put energy into if this is an important effort to you.

Carolyn Woodard

And I don't know, controversial maybe, but you could use AI to help you organize and uh develop a marketing campaign or uh you know help bring um local groups together to uh understand where the data centers are going in and how they can influence them. Those are all things that AI can help you with. So this kind of circular circle of uh using AI to fight the data centers potentially. Uh

Carolyn Woodard

Both Google and Microsoft have pledged to be water positive by 2030. That's only four years away. So uh it's not clear how transparent they're being about what water positive means, but it's certainly something that they are feeling a backlash on. And also, like I said, they have a profit motive. So it's in their interest to use water efficiently and in a low-cost way, uh, especially in communities where it's a very high cost.

Carolyn Woodard

So I guess my takeaway would be I'll share some resources with you on how, well, not how electrons work and why it gets so hot when you when your computer is computing things, but um just that uh making informed choices about using AI, I think, is the smart way to go. So becoming AI literate and then also focusing on maybe local advocacy, ask your vendors about their commitments to clean energy and to water usage, um, supporting clean grid advocacy in the region where you're working or nationally, uh, wherever data centers are operating. Um, that is

Carolyn Woodard

Those are steps that you can take that are active that don't involve just deciding to not use AI, uh, which is kind of keeping yourself from using these very valuable tools. Uh, and whether that hurts yourself more or hurts the company more is open to debate. So getting involved in that advocacy around the data centers, I think, is a very uh good way to have agency.

Carolyn Woodard

Another article and idea I wanted to bring to you is this article that I saw in the Atlantic about uh the AI bubble and uh what that means for nonprofits. And I have said that nonprofits tend to be very cautious about technology and new technologies. So we kind of never want to be on the bleeding edge of a technology. That's not where most nonprofits feel very comfortable because it's a commitment and an investment. And when you're on the bleeding edge, they say bleeding for a reason, it it's much more risky, it's more likely to fail. And before the market has really settled down, you may be investing in a company that won't be there six months from now.

Carolyn Woodard

So while consolidation is happening and while uh tools are you know evolving so quickly, in some ways it's right for nonprofits to be cautious and wait for the market to settle down. So six months ago, a year ago, I think I was hearing from a lot of nonprofits and at conferences and online that a lot of nonprofits were really just waiting and may have even been a little bit skeptical. Is this just a flash in the pan that I'm not even gonna need to know about? Like it's a cute parlor trick you can ask ChatGPT about how to date women or something like that. But it's not gonna be really impactful in our work, in either how we do our work or the work that we're doing as nonprofits. And

Carolyn Woodard

I think a year later, that horse has pretty much left the barn. AI tools, even if you never use a standalone AI tools, they are being incorporated into every other tool that you already use. Zoom, MailChimp, your website, SEO, your CRM, your ERP, QuickBooks, uh, Asana, productivity, like everything that you're using, every time you update it, you'll see there's a new AI chat, there's a new AI helper, there's a new AI icon that you can click on to use AI with our tool. So it really is going to be and is already almost everywhere that you're working. So I don't think nonprofits can really say this is just a flash in the pan that I'm not going to need to know about. Uh especially the productivity gains. In the past year, they've become really measurable. Uh, you can complete your tasks uh so much quicker. It's transforming whole departments at your nonprofit for sure.

Carolyn Woodard

If you talk to your comms and marketing teams, uh they are really aware of what AI can do and what they're doing with AI. Uh, same with probably your database people, uh, if you have developers, um, all of those people are really interested and are already, well, as we said, everyone is probably using an AI tool. The different people and the different teams are not kind of universally using the same tools to do the same things, of course.

Carolyn Woodard

With limited budgets, limited IT staff, uh, often a kind of limited IT management capacity, uh, maybe a limited leadership around IT as a strategic investment at your nonprofit. Of course, it's totally reasonable to be cautious. Um,

Carolyn Woodard

As we're learning, AI seems to be very easy to adopt, but it really has a lot of change management kind of built into it. It has a lot of front-loaded learning that you and your staff need to do. So the knowledge work that nonprofits do every day, communications, grant writing, program design, evaluation, case management, all of those are really kind of in a sweet spot for AI tools to help with. And some aspects of them are very automatable. Uh, so that's something that you can either ask AI to do or build agents that can do that busy work for you.

Carolyn Woodard

AI is also extremely useful in recognizing patterns. So something we've been thinking about and talking about on this podcast as well is uh using AI tools to look over your historical data as a nonprofit, maybe your program data, um, your your clients, your constituents, your donors, all of those things that help you see patterns that you haven't been able to put together. So that's a really good way to use AI.

Carolyn Woodard

I guess what the Atlantic article says to me about the consolidation that's happening around some of these larger tools like Claude and Perplexity and Gemini and Copilot is that in the past you might have had to wait a couple of years or several years for the market to settle down.

Carolyn Woodard

I remember in the early days of email marketing, there were, you know, 20 different companies that were all available to nonprofits. And then it really consolidated into a solid two or three. That took several years, though, for that to happen.

Carolyn Woodard

With everything being in in hyperdrive with AI, it is happening much more quickly. Uh so I'll share that link to the Atlantic article. It is behind a paywall, unless you have a subscription or I think there's several free articles per month that you can see.

Carolyn Woodard

But I think it's more important for nonprofits to re reframe that question of should we adopt AI? Should we invest in AI? Should we wait? To, you know, what's the cost of waiting at this point a year or two years into it? And what's a responsible first step? And as you know from our webinars and this podcast, you know, policy, governance, talking about it, talking about it with your staff, with your leadership team, developing some policies, keeping the conversation going as AI changes so quickly, you know, have regular check-ins on it. If you have a monthly all-staff meeting, set aside some time every month to just check in on AI, how everyone's feeling about it, what they're using it for, and uh where your policies are and may need to change.

Carolyn Woodard

Another thing I want you to take away from this idea that the market really is consolidating is that now is the time to get into some of those tools. As we always say, log in with your work email, purchase it if you are able to, because then you get those uh business protections with the terms and conditions.

Carolyn Woodard

Identify a couple of low stakes places where you can experiment. Maybe you have some early adopters, maybe you have some teams like your comms team or your database team, your grants writing team that are already interested in experimenting with it and uh build in some low stakes, as I said, low-risk places that you can pilot and learn what you can from that pilot and then maybe expand farther.

Carolyn Woodard

So, not a bubble. AI is not a bubble at this point. Now, it's that I don't want to say like my crystal ball, like there's not going to be any kind of crash with the stock of those tech companies because as we know, lots of things can happen and can happen very quickly in this space. So I guess watch this space, watch the stock market, watch the news. Um, but it's it's not going to go away. It's going to continue to influence how we work and uh just the sheer volume of the data centers that they're building out is also another indication of that. So,

Carolyn Woodard

Final thing I wanted to share with you is you may already be aware, I hope you're aware, uh, Fast Forward's 2025 AI for Humanity report. We talked about it in this podcast. I'll share the link to it. They surveyed 200 nonprofits across lots of different budget sizes. They found that 82% of nonprofits were already using AI for internal operations.

Carolyn Woodard

They found that those using off-the-shelf tools were less likely to have AI policies or risk controls in place. So, this is one of the motivations for us always saying, you know, purchase that license if you know it's in your budget to do so because it's um more likely to lead you to be talking about it with your leadership, developing those policies, and you're more protected again if you have those uh paid licenses in place. Uh

Carolyn Woodard

84% of nonprofits in last year's report said funding was their biggest barrier to scaling that AI work. So again, if you try a pilot project or two, and you could show that to your funders and say, you know, we need to train all our staff to be able to do this, or we need to be able to train the entire team to be able to do it. That's um that's more convincing, of course.

Carolyn Woodard

But the opposite side aspect for all the funders listening is that um a lot of funders are upping their game, but need to up their game around training and productivity at their grantees. Uh, not just ways that you can use AI for your mission, but ways you can use your AI for your work and how to make sure that everyone on your staff is AI literate and that all that change management is happening in the way we are working. So your staff, some of your staff members aren't members aren't left behind.

Carolyn Woodard

So the 2026 report is currently in progress. They have opened the survey and they are recruiting survey participants now. So this data is going to shape how funders, policymakers, and tech companies understand nonprofit needs in this sector for AI. So

Carolyn Woodard

If you are a nonprofit founder, executive director, CTO, interested board member who knows most about how your organization is using AI, this is something that is for you. Please take this survey. I will include the link in the show notes. And if you're on LinkedIn, you've probably seen it there. It's being spread widely.

Carolyn Woodard

Please participate. The more people participate, the better their survey will be.

Carolyn Woodard

And in addition, we've talked a lot about agency in this podcast and how at this moment the philanthropy and nonprofit sectors are really needed to talk a lot about what we need, how we want these tools to look, how safe we want them to be, what we want them to be used for, how we want these companies to behave.

Carolyn Woodard

All of that is really important to have the data, do the analysis, do the evaluation, as we talk about a lot in nonprofits. So please take that survey if you are able to. If that's you, if you're the person in your organization who knows the most about how you're using AI, please take, you know, five minutes, 10 minutes to take that survey and help this report be accurate about what we want and what would really help us.

Carolyn Woodard

So just in closing, um, three points. The environmental concerns are really real. They're totally real. You are not crazy to be concerned about them, but I don't want everyone to kind of direct themselves into refusing AI or criticizing AI when there's also the opportunity to focus on that advocacy for a cleaner grid, which we're gonna need anyway, by 2030. So getting on it now, uh holding those vendors accountable, making sure they understand how important the environmental impacts of these tools are to you, to the consumers, is really, really important. So take that into consideration while you're also considering, at least make your decisions from an informed uh viewpoint. So get AI literate before you decide not to use AI. Uh

Carolyn Woodard

Productivity case, as this market consolidates, as it appears that it's less and less of a bubble and more and more of a movement that's here to stay. Uh it's harder and harder. I know there are a lot of nonprofits probably listening that are cautiously dipping a toe in, maybe not fully in AI policies yet, haven't developed that governance, really using AI just in a very ad hoc manner to your nonprofit. There's no blame, there's no judgment. We listen, but we don't judge. Uh, but getting in now, understanding the tools, maybe doing a pilot, talking among your staff and with your leadership about it. Uh, it's

Carolyn Woodard

It's something that the longer you hold off on having those conversations, yes, the market will be more and more consolidated, but you are getting to a point where you're going to have missed those productivity gains. So, and that's gonna start to become something that your funders are gonna be asking you about, your donors are gonna be asking you about, your staff are gonna be asking you about. So, where's that policy? I want to use this AI tool, and I'm not sure how I'm supposed to at my organization. So, making sure you're having those conversations, the longer you wait, the more behind the curve those conversations are gonna be.

Carolyn Woodard

And then again, participating in what comes next, participating in shaping it, take the fast-forward survey if you're the right person at your organization. And I'm really looking forward, I think we're all just waiting to see what the 2026 report says. And that is gonna be in so much use throughout the industry, the sector. So, getting in and having your say at this point is another good way to have agency in these big, big questions that are impacting us now with nonprofits in AI. So, I will share all these resources with you in the show notes.

Carolyn Woodard

Thank you so much for listening. I'll be back here next Tuesday with more nonprofit in AI news and resources, and of course, on Friday with our regular technology topics podcast. So, until then, take care.