Community IT Innovators Nonprofit Technology Topics

Nonprofit AI: Environmental Costs, New Philanthropy Dollars

• Community IT Innovators • Season 7 • Episode 40

Share episode

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

0:00 | 27:05

Carolyn Woodard covers the real-world costs of AI infrastructure, a coming wave of AI-generated philanthropic wealth, and how to think about AI fundraising tools this week.

The headlines about AI tend to focus on what it can do. This episode looks at what it costs — in water, electricity, and rate increases that are already hitting low-income households — and at the enormous wealth being created on the other side of that ledger. 

Carolyn also responds to a listener question about AI tools in fundraising, drawing a comparison to the early days of social media for nonprofits and why the same change management instincts should apply.

This episode covers:

  • Data centers in Georgia and Arizona have drawn water without authorization, and projections suggest Texas data centers alone could draw down Lake Mead by 16 feet annually by 2030. MIT researchers found that AI-specific energy use could equal 22% of all US household electricity consumption by 2028, and that utility deals with major tech companies are already shifting infrastructure costs to ratepayers.
  • Nan Ransohoff's widely discussed Substack piece argues that AI wealth creation could generate $37 to $100 billion annually in new philanthropic capital. Forbes counts 45 new AI billionaires in 2026 with combined wealth of $2.9 trillion. Tech-fluent nonprofits are likely to be better positioned to build relationships with this new wave of funders.
  • AI fundraising tools are at a moment similar to early social media: some organizations will jump in, some will wait, and neither is automatically right. The change management skills your organization has built through past fundraising shifts can apply here. Just because the tools are new, don't think you don't have the leadership to manage the change.
  • Board.dev connects nonprofits with tech-savvy board candidates and offers 28 AI governance questions your board can use right now.

Resources Mentioned:

  • Fortune, "America's Data Centers Are Thirsty. Rural Towns Are Paying the Price" — https://fortune.com/2026/05/13/data-center-georgia-arizona-water-wars/
  • MIT Technology Review, "We Did the Math on AI's Energy Footprint" — https://www.technologyreview.com/2025/05/20/1116327/ai-energy-usage-climate-footprint-big-tech/
  • Nan Ransohoff, "The Third Wave of American Philanthropy" — https://nanransohoff.substack.com/p/the-third-wave-of-american-philanthropy
  • Forbes, "Meet the 45 AI Newcomers to Forbes 2026 Billionaires List" — https://www.forbes.com/sites/phoebeliu/2026/03/10/meet-the-45-ai-newcomers-to-forbes-2026-billionaires-list/
  • Board.dev — https://board.dev

New every Tuesday.

_______________________________
Start a conversation :)

Thanks for listening. 


Carolyn Woodard

Hello and welcome to the Community IT Innovators Nonprofit AI Podcast Midweek Check-in. I'm Carolyn Woodard, your host. I'm not an AI expert. I am just here to bring you some news, some resources, some of the conversations that are going on right now around AI and nonprofits with implementation, new tools, new ways of working, and try to all get smart together about these kind of amazing new tools, but they also have a lot of issues bound up in them, we can say. So

Carolyn Woodard

Today's episode is about some real costs of AI to the planet and some implications for philanthropy. I wanted to share a couple of resources with you, some recent news articles that I just found interesting to sit and think about, and some new information to add into some of the things we've already talked about, especially with regard to the environment. So

Carolyn Woodard

There were two data center incidents that were written up in Fortune in May and that kind of caught my eye. They were both about taking water without authorization. So in Georgia, there's a data center developer called QTS. I feel like this is one of those things that's happening too, as with supply chains, where the main company can say, oh, it was a subcontractor of a subcontractor. We can't possibly be responsible for their labor practices or their environmental use or what have you. So in the data center debate as well, the company, the AI company doesn't make the data centers, they contract, and then the construction company is a contract, and then they have subcontracts, et cetera.

Carolyn Woodard

In Georgia, the data center developer used 30 million gallons of water through unauthorized connections. That's 44 Olympic-sized swimming pools, and it caused low pressure in a pretty affluent suburb that noticed and went to the city council to find out what was going on, and it definitely exceeded the limit that had been agreed to during the planning process. And at the same time, I think both Georgia, I did not realize, was under a drought, but uh Arizona also has been under drought conditions for several years now. So they have really serious water usage. You know, they're keeping an eye on things. Um,

Carolyn Woodard

But they had a similar problem where the contractor was um uh using the water that they were not allowed to be using.

Carolyn Woodard

The scale is really important to think about. In Texas alone, there's a statistic that data centers could consume as much as drawing down Lake Mead by 16 feet in a single year by 2030. And Texas is already in a water crisis. Um, so this is leading to cities enacting bans or moratoria on new data center construction. Some states are trying to put in more regulation.

Carolyn Woodard

I've said in the past on this podcast, this is really, if this is important to you, you can get involved locally and show up at those meetings and understand what's going in with the data center siting and tax breaks and uh electricity use and water use. So those are all very good things that your voice can make a difference in.

Carolyn Woodard

And in regards to this kind of developing story around the energy use and the water use, there was a new MIT ebook that has come out. MIT did uh several months doing research and working the math on AI's total resource footprint. It's a long article. I'll include the link for you to have a look at it. Uh, it is behind a paywall, but you're allowed a certain number of free articles per month. Um, so you can hopefully we'll be able to have a look at it. It's a very dense uh PDF, which goes into a lot of the science and a lot of the calculations.

Carolyn Woodard

So if that's something that you know you're interested in or just want to get a really reputable source to tell you the story, uh several things jumped out at me from this long article. Um,

Carolyn Woodard

One was about the AI energy cost per query, which is this statistic that was used a couple of years ago. If you remember, if you make one request, one prompt to AI, it's like pouring out a bottle of water. Um, which this article talks about is really not the right way to frame it. A text query, depending on how far you are from a data center, uses about the same amount of energy as running a microwave for a few seconds. So, you know, all households, almost all households use microwaves. We wouldn't think about like, oh, I shouldn't microwave my hot pocket uh because it's using energy.

Carolyn Woodard

So this article talks about trying to reframe that. So we do have to take into account that, you know, so many of us are making multiple queries every day. Uh, every query, there's hundreds of millions of users. I mean, we know ChatGPT was one of the highest, fastest growing downloads uh from the App Store.

Carolyn Woodard

This article tries to make the case that by 2028, AI-specific energy use in the US could equal 22% of all US household electricity consumption annually. The share of US electricity going to data centers is expected to triple from 4.4% to 12%.

Carolyn Woodard

So that's a better frame for the AI energy use than per query or per person.

Carolyn Woodard

The electricity data centers already use is already 48% more carbon intensive than the US average, partially because so many are concentrated in places like Virginia in Northern Virginia that rely heavily on fossil fuels, both coal-fired-fired plants and natural gas plants.

Carolyn Woodard

And this cost may not stay abstract. Research suggests from MIT that utility deals with big tech, the five, six big AI companies, can shift infrastructure costs to ratepayers, which we're already seeing in Virginia. One Virginia estimate put it at $37.50 per month extra for residential consumers of electricity because of the way the grid is structured.

Carolyn Woodard

The transparency problem runs across both the water and the energy usage. Companies treat these figures as trade secrets. Researchers at MIT called it a total black box. So they are making inferences and using the best data and the best information that they can based on what's already happened. And uh, but they they can't get the actual data because it's private companies and that is secret.

Carolyn Woodard

If your mission touches on environmental issues, energy equity, housing, water access, low-income communities, this is not a future tech story. This you know that this is happening already. I have heard of people whose energy electricity bills have gone up 900%, 1,000% over the past year. And a lot of states are struggling because they may have programs that keep your electricity from being turned off in the middle of a heat wave or your heat from being turned off in the middle of the winter.

Carolyn Woodard

But those programs' funding was not taking into account these types of rate increases, and so they're scrambling right now to try and protect those low-income users from losing their air conditioning on a hundred-degree day.

Carolyn Woodard

If you are interested in this as a public health story, community rights story, land use story, and you're evaluating your AI tools based on your organizational values, this lack of transparency from these major vendors, these um issues of water usage and electricity usage. You know,

Carolyn Woodard

I know I've said in the past podcast that it's in these companies' interests to convert to sustainable, less expensive uh energy and water over time. So maybe in the long run, it's going to be beneficial to have these companies who are trying to pay less for their energy and water and the land use eventually to be involved. But that impulse to pay less, I mean, they're capitalists and they're looking at their profits and they're looking at their uh financial picture right now, which is leading them to quote unquote steal this water, uh, leading them to use electricity under existing guidelines, and uh hope that the state doesn't realize how much electricity they're actually using and what the rates, how high and how quickly the rates are going up.

Carolyn Woodard

So if you're worried about the environmental impact, you're not wrong to be worried about it in the short term. There are still lots of things that you can do locally and around advocacy that can help those companies become more regulated. And

Carolyn Woodard

To me, it's a little bit similar to the framing around recycling that all of recycling was pushed onto the end consumer. Like you should make sure that you recycle all your plastic. And it's like, well, what about the company should be making sure that it doesn't make products that have to be recycled because we know that only a fraction of those water bottles get recycled?

Carolyn Woodard

It's the same question here. We shouldn't be accepting this push down to the consumer to just use AI less. We should be demanding that those AI companies are paying for their own electricity and using water in closed systems. Uh, the Arizona um data center is planning to use a cool closed system where they won't be taking any more water out of the environment. It'll all be in this, these pipes that just cycle around. Well, that's five years from now, is when that will be fully operational. So they'll be using the water in the meantime. Thank you very much.

Carolyn Woodard

So something to keep an eye on, especially locally. If you have data centers in your local area, you know, get involved at that level. And that is a good use of your advocacy time. So

Carolyn Woodard

I want to move on to kind of a flip side of this, is in addition to the costs, environmental and uh energy, there's also this just phenomenal, we're in this bubble right now of funding and valuations and IPOs and you know people at these companies who are, you know, have shares and are having their shares missed, uh, and investors in these companies as well through venture capitalism. So there's a lot of money being made right now on the other side of this question. So

Carolyn Woodard

I wanted to share an article with you that's kind of been controversial, getting a lot of pushback in LinkedIn, getting talked a lot about on LinkedIn and at foundations, um, is this article called The Third Wave of American Philanthropy, which was published on Substack by Nan Ransohoff.

Carolyn Woodard

And her argument is that AI is about to generate an enormous wave of new philanthropic capital. So potentially 37 to 100 billion per year in new philanthropic spend. And this is, she talks in this article about it's basically back of the envelope math. So she's just estimating a lot and kind of calculating out based on equity stakes, uh, giving pledges from OpenAI Foundation and from anthropic employees. And

Carolyn Woodard

The tone of this piece is very, she's very rooted in Silicon Valley and kind of critical of existing philanthropic structures. And perhaps there's an argument to be made there about the ability to absorb that amount of uh funding coming in and that maybe new structures are warranted, the old structures maybe aren't great to begin with, and maybe this is going to cause a crisis of being able to get that money in and then out to the recipients for sure.

Carolyn Woodard

But she also talks a lot about philanthropic startups. Like, there's no way that the philanthropy sector could is even maybe the right sector to be doing this transfer of wealth from the donors to the recipients, and um, that those foundations need quote unquote tech caliber talent, that the donors are going to require really smart people running these, if it if it's a family foundation, if it's a donor-advised fund, uh, if it's a donation to an existing foundation, um, that there'll just be, I mean,

Carolyn Woodard

I think she's not wrong to observe that there will be lots of skepticism and new people coming into philanthropy, especially from tech, you know, self-made billionaires, are gonna have a lot of ideas about how better talent at the top could help with distribution or new ideas, new disruptive ideas. They come from this disruptive tech environment, so they're gonna want to disrupt philanthropy too. Um,

Carolyn Woodard

It's rubbed a lot of people in philanthropy and traditional nonprofits and foundations the wrong way. It's understandable that people are giving a lot of pushback. Um,

Carolyn Woodard

But the underlying data, I think, is fair. Forbes has an article on the 45 new AI billionaires in 2026. They have a combined wealth of 2.9 trillion. This is the fastest wealth creation in tech history.

Carolyn Woodard

Even if the bubble bursts, even if a fraction of those individuals become philanthropists, even if they do so at rates that are not the same as in other industries or in other waves of money coming into philanthropy, it's still gonna be, even if it's just a fraction of that, it's gonna be a lot of money.

Carolyn Woodard

And the practical questions for nonprofits is that new money tends to go where there are relationships to organizations that they already know, or they start their own new organization that's gonna be better in their eyes. Um, so we want to think about um this wave of funders. They have different values, they have different metrics, different expectations. It's not a reason to panic. I've talked in the past about there seem to be these very large AI grants coming online, like from Google, uh, from Microsoft. So it's a reason to pay attention.

Carolyn Woodard

One of the implicit arguments in this Ransa hoff piece is that tech-fluent organizations will be better positioned to build up those relationships with new philanthropists and attract this new money, and they'll be better able to use it, right? So that's we're sitting with and thinking about whether your organization and your board reflect the kind of technology fluency that will help you navigate this new landscape as this new money comes into the system.

Carolyn Woodard

I want to recommend a friend of the pod, friend of the organization, Board.dev. They are a matchmaking organization that connects nonprofits with tech savvy individuals who want to serve on nonprofit boards. And their research has found that this is one of this is a huge driving factor in both the nonprofits' ability to successfully manage their IT in general and this change management to AI tools specifically. So

Carolyn Woodard

Having a tech -savvy board member is a real lever that you can use. And Board.dev is set up to help you recruit uh new board members that have a tech background, and it also works on with those individuals to help with training of how nonprofits work and how a nonprofit board works, and like what you could expect, and how quickly change can happen, and how you can be most helpful to the nonprofit that you appreciate enough that you're gonna serve on their board.

Carolyn Woodard

So they're just a phenomenal organization. I really recommend them. Link will be in the notes. But if your board doesn't have a tech savvy individual right now, it's worth having a look at these resources. They also have uh on their website 28 questions for your board members to talk about with regards to AI. So they just have a ton of great resources, and I really recommend that. Uh,

Carolyn Woodard

It is important to note that all of this is paper valuations tied to the private market and venture capital. Uh, if the IPOs don't go well, if the bubble deflates or bursts, the projections are going to shrink too, but it's worth paying attention to.

Carolyn Woodard

Some of the numbers from this Substack article. The OpenAI Foundation holds 26% of OpenAI, which is worth about $220 billion today. Anthropic 7 co-founders have pledged to give away 80% of their wealth. They're instituting an aggressive donor matching program for their employees.

Carolyn Woodard

And so all of that leads into, as I said, her back of the napkin math to kind of evaluate how much that's going to be. She says, for example, if it turns out to be about $50 billion a year, that alone would fund, fully fund the annual budgets of six Gates Foundations.

Carolyn Woodard

She talks about the three waves, the early philanthropists like Carnegie, and uh then the second wave after you know the growth of Microsoft. Basically, when Gates Foundation was created, I think all of us were like, wow, they've got a lot of money to give away. And they are very interested in doing a lot of global philanthropy, and they've really moved a lot of the sectors.

Carolyn Woodard

So this is um, you know, kind of that when Gates Foundation was created on steroids. So that's a way to think about the amount of money that's gonna come in. Uh, and

Carolyn Woodard

I just do recommend reading this article as very interesting ideas and kind of the scale of what's happening and uh what we have at the moment and how it could absorb that amount of funding. And also, of course, just this question about what that funding is going to go to.

Carolyn Woodard

And if you are in the business of trying to understand what that needs are and matching that up with funding, it's just a really worthwhile article to take a look at.

Carolyn Woodard

And then I wanted to just pivot a little bit and um kind of this question came in around AI fundraising tools.

Carolyn Woodard

So there's new platforms, there's AI for donor research, donor outreach, AI for grant writing. And if your organization is stretched thin and worried about using AI tools in development, um, this question really reminds me of the early days of social media for nonprofits.

Carolyn Woodard

There were some organizations that jumped in and went viral on Facebook. You know, if you think about the um Ice Bucket Challenge, but just other issues too, of um there were lots of stories in that day about using Google ad grants, for example, and getting, you know, 600% increase in kitten adoptions or just different stories like that about using the social media platforms and realizing that people on social media really cared about kittens and about adoptions and about feel-good stories and about the things that nonprofits were doing.

Carolyn Woodard

And so it was a natural place for nonprofits to have a presence. Some of those organizations, of course, went viral and then it was very difficult to handle that virality. There were organizations that made mistakes and had reputational consequences from not really thinking through their social media position. There were organizations that waited and maybe got into social media very late. They kind of missed some of the what they could have been, you know, interest they could have been attracting or donations that they could have been attracting.

Carolyn Woodard

And then there were some that made thoughtful decisions, you know, and were kind of intentional about getting involved in social media and which platform was a really good fit for them. And you just see, I think now it's kind of settled down and there's best practices for the different platforms and for presenting yourself there. But none of those approaches in the very early days were automatically right or wrong. It depends on your organization, your mission, and of course, leadership. So

Carolyn Woodard

AI and fundraising feels like it's at a similar point right now. And here's why the I think the answer is you, of course, you should be paying attention whether or not you're ready to jump in. Because fundraising is one of the most relationship dependant and trust-sensitive functions at any nonprofit. And it's, of course, it's very foundational to the existence of your nonprofit, that funding.

Carolyn Woodard

So if you use AI tools in ways that feel impersonal, that violate your donors' expectations, that your board or leadership isn't aware of or hasn't bought into, there are real harms that can be there. And

Carolyn Woodard

The same change management principles that apply to any major shift in development operations, I think just should apply here.

Carolyn Woodard

So I think my advice would be get educated, self-educated, get AI literate, use AI tools yourself for other tasks to begin with, you know, for writing. So you understand how they work, what they can do, where they have gaps, where they cause, could cause problems, how much you have to edit, how much you cannot trust that just putting something that an AI has written directly into your email and sending it out, right?

Carolyn Woodard

It's important to have conversations at your organization now, before there's a lot of pressure to make fast decisions about those AI tools. Develop a policy, at least a set of guiding principles that reflect your values, and be transparent internally about where you have uncertainty about what you are doing, should be doing, and can do with AI tools in fundraising.

Carolyn Woodard

If you've ever switched CRMs, if you've changed your gift processing system in the past, if you overhauled your major donor program, if you pivoted to a different type of event gala or donor ask, you've already done this kind of change management. It's just AI tools. The change management skills are the same, being intentional about it and understanding what the tool does and what the user journey, you know, the donor journey is on the other end of your communication with them or your appeals, your, you know, how you're doing your research, all of these things.

Carolyn Woodard

So you probably have more experience than you think in change management around fundraising. So the fact that this technology is new, it's not immaterial, but your leadership skills to manage a thoughtful transition in fundraising tools is already there. So it's just a new tool, it has far-reaching implications, and uh it's maybe more difficult to adapt than we've been led to believe, but I'm sure that you have change management skills from past changes that you can apply in this situation. So uh

Carolyn Woodard

That's all I have for you today. I will share these resources in the show notes. Um,

Carolyn Woodard

We will be back with you on Friday with uh part one of our podcast about the webinar, which is Wednesday. If you're interested in coming to that webinar Wednesday, I will be talking with our senior consultant, Nuradeen Aboki, about some case studies from his work over the past year in working with clients on assessments, intentional adoption of AI tools, and that change management I was just talking about. So he's got some really good experience and recent experiences that he's gonna share with us. So it's not just theoretical. He's really gonna talk about some of our clients that had some success, maybe some of the things they wish they had done differently. And so I think it's gonna be really valuable.

Carolyn Woodard

So if you can come to that webinar, it's this Wednesday. And if you miss it, you can always catch the video on our website, community it.com, or there'll be a two part podcast as well this Friday, next Friday, covering the same topic. So I hope you can join us for that. If not, or if you are just waiting for nonprofit AI, I will be back in your feed next Tuesday. Until then, take care.