Setting a 'Tech Agenda' for Climate Week

Audio of this conversation is available via your favorite podcast service.

​From September 21–28, New York City will host Climate Week. Leaders from business, politics, academia, and civil society will gather to share ideas and develop strategies to address the climate crisis.

​The tech industry intersects with climate concerns in a number of ways, not least of which is through its own growing demand for natural resources and energy, particularly to power data centers. What should a “tech agenda” for Climate Week include? What are the most important issues that need attention, and how should challenges and opportunities be framed?

​Last week, Tech Policy Press hosted a live recording of The Tech Policy Press Podcast to get at these questions and more. I was joined by three expert guests.

• ​Alix Dunn, Founder and CEO of The Maybe
• ​Tamara Kneese, Director of Data & Society's Climate, Technology, and Justice Program
• ​Holly Alpine, Co-Founder, Enabled Emissions Campaign

What follows is a lightly edited transcript of the discussion.

Justin Hendrix:

I am so excited to have the three of you here today, and we hope to have a wide-ranging conversation. I hope, also, that it'll be a conversation between us, meaning that you'll each take the opportunity to respond to one another and to build on things that you're hearing, to challenge things you're hearing and to add nuance wherever you see fit through this conversation. But it felt to me a good moment to step back and to think about some of the threads that I know each of you are working on very intently in your research and your activism.

We've got Climate Week coming up where I live in New York City next week, September 21st through the 28th. A lot of folks are going to be in town for conversations around climate-related issues. We'll have heads of state, we'll have various government ministers and regulators and even a few celebrities, I understand, who are participating in the different discussions. And then of course, this is all at the outset of a UN General Assembly in town, various other kind of events that happen around that. There's just lots going on in New York.

And this issue around the sort of intersection of tech and environment tech and kind of concerns about climate change has really reached a crescendo in the last few weeks and months, particularly I think given how much curiosity there's been around AI infrastructure and data center development and just the extraordinary amount of money that's being spent there. But that's not the only issue, of course, that that intersection. And I'm hoping we can get a little sense of how each of you see these questions as we go along.

But Alix, I think I'm going to come to you first and maybe give us a minute or two, the elevator speech on The Maybe on what you get up to there. The community you're building of expert storytellers, media organizers, et cetera, working on tech politics. But also if you could maybe set the stage for us why Climate Week is important, why you're coming to New York next week, for instance, to be part of it.

Alix Dunn:

Yeah, happy to. I think the kind of elevator pitch for what we're doing at The Maybe is to support the extremely diverse and global community of people that know a lot about how technology's reshaping our political environments and aren't necessarily listened to in mainstream conversations. Partly because I think a lot of times we speak to each other, rather than to broader publics. So trying to encourage and support people that do have this knowledge to be more effective communicators on the global stage. So we run the New Protagonist network, which is partly a media training program, partly a community of practice of people that are trying to shift the narrative of AI politics particularly. But technology, politics and AI is like, I don't know, it's all kind of same for same.

And we have done some work in the last year and a half on the rapid expansion of physical infrastructure, of computational facilities, of big tech companies. Particularly a piece of work that we released this spring looking at five different countries where data centers have been built out, how that happened. Did it happen with local consultation, good democratic process, good consideration of the environmental impacts of these centers? Spoiler alert, no. But we wanted to sort of get into these specific examples so people could see that this isn't a US problem, it's not just any single country problem. This is a global dynamic where big tech is essentially rapidly investing in environmentally devastating projects globally as part of a larger project to push a vision of our future on us, I think.

For Climate Week, it felt like a really lovely moment where lots of people were going to be converging in the same place and we try and figure out where's a good place to have a particular conversation. And a few months ago decided Climate Week might be one of those. I will say I find it a bit ironic and strange that it's in New York, given what's happening in the US right now in terms of turning the collective backs away from doing meaningful work on improving the sort of trajectory with climate. And also obviously the rising dictatorship, I would say, in the US.

So I'm feeling a bit mixed about showing up in New York and whether or not it's still the place we should be having these conversations. But I'm excited to see, I think all of you and I'm sure lots of people that are dialing in. So we're excited to be there to be a part of the conversation and hopefully to host some good discussion and community building with people that have been working in solidarity with each other for a long time, and don't necessarily get the face time. So that's our plan for next week, and we are going to be, I don't know, I'd love to see you guys there.

Justin Hendrix:

I'm going to come back to you on some of the specifics of what you've learned in your research, in particular. I have also in just recent days heard from individuals who are really having a hard time getting visas unfortunately to travel to New York for the UN General Assembly and other events. And so it's certainly a concern whether those types of activities are going to go off as they have in past, or whether that discourse will be stunted to some extent.

But Holly, I want to come to you and learn a little bit about your elevator pitch for the Enabled Emissions Campaign. And if you will also the basics of your story, how you came to this activism. I think our listeners would benefit from knowing a little more about it.

Holly Alpine:

Sure, thanks. And I took one of Alix's trainings before and it was great, so can give a big plus one to that. So thank you for leading those Ali. I spent a decade at Microsoft where I helped launch various sustainability initiatives across the company from starting the first global data center community sustainability program. So I was on Microsoft's data center team for many years and I also co-founded and ran Microsoft's employee engagement program for sustainability. So got to interact with hundreds of thousands of employees who really care about sustainability all across the company.

And saw that Microsoft and other tech companies have some pretty amazing pledges for minimizing their own environmental footprint. Of course they can do more, but I was there during the rise of Microsoft's increasing commitments around sustainability, very strong pledges around carbon water waste and ecosystems. Very strong pledges around how technology should be used for good, not ill and was really proud of working at the company.

But I also saw up close how this company that I loved and who was doing great work on minimizing their own environmental footprint, was also creating and deploying technologies for big oil to dramatically increase global fossil fuel production. Now, this was technology that Microsoft engineers were working with these companies to create that was then being used by big oil in various capacities to optimize their production or their operations, to increase oil and gas production.

And we saw, so we did some math there and looked at some figures and saw it was astounding how much emissions were coming from the production due to this technology. And so we ran an internal campaign for several years, internal activism and organizing. Really felt like we reached the limits of what internal activism could do if there was no external force. Which at the time we really didn't find any external groups who were pushing to increase accountability for the technology that these companies were creating. They were just having no accountability for how their tech was being used.

And so my partner Will and I decided to leave the company. Very hard decision because we loved our jobs, we loved the company, loved our colleagues and team, but we decided this was something we needed to do. So we left in January of 2024 and co-founded the Enabled Emissions Campaign. And our mission is to call attention to this overlooked yet massive driver of climate risk, and push for accountability in the tech sector for Microsoft and beyond.

Microsoft is the largest provider by far more than all the other providers combined, but we know this needs to be a tech-wide issue and needs to be one that we talk about at places like Climate Week. We're seeing a lot of conversations around AI and climate, but most of it revolves around its infrastructure, which is definitely important. Those impacts should absolutely get addressed. But what we're not even seeing at all is what the tech is used for, and how it actually fuels more, but also fuel production. So I will be at Climate Week next week and hope to have a lot of conversations about that.

Justin Hendrix:

I do want to figure out how to come back to that question of how you think about the broad array of ways that the tech firms are involved in climate change phenomena. So I'll ask you maybe to kind of dig more into that as we go through the conversation.

But Tamara, I want to come to you and the quick elevator pitch on the Climate Technology and Justice program line of work at Data & Society and why you're coming to Climate Week.

Tamara Kneese:

It's really great to be here with all of you and obviously we all know each other, which is really nice. And I would say that what I'm really interested in doing at Climate Week in particular is cutting through a lot of the corporate narratives that we hear over and over again about AI and why we're gunning for it, and also why our government is investing so heavily in it right now, and a lot of the speculative promises attached to it. So thinking about yes, AI will help solve climate change without any details ever pointed to in that respect.

And I'm having also worked at a large tech company on a sustainability team, I know what the internal processes of gathering data around environmental impacts can look like. But then having done a lot of work as an organizer and also as a researcher who focuses on participatory methods, and really engaging the communities who are most impacted by AI systems and algorithmic harms, but also by the materialities undergirding AI.

So really thinking about how AI's larger infrastructures across the supply chain. And I think data centers have become this really interesting focal point, a space where we really see a lot of forms of resistance, where we see a massive amount of both corporate and government investment and speculation right now. And they also have the capacity to really become sites of resistance for communities at particular sites all over the world.

And at the same time, AI is very much dependent on these much larger supply chains. And I think that's why Holly and Will's work is also really important, thinking about what AI is actually being used for. So thinking about the applications of AI to a lot of harmful practices, including military practices and genocide along with obviously perpetuating and exacerbating the fossil fuel industry.

And so what we're really trying to do in the CTJ program is form a community of practice and really get people who are coming from the more computer science side of things, who are industry practitioners too, who are maybe also organizers and activists who know how those internal metrics come about. And can cut through some of the bullshit that we're often fed from tech companies who are like, don't worry about it. It's just like a tablespoon of water per prompt, to really think about those larger supply chains and the larger infrastructures that are obviously really affecting communities on the ground.

And so part of the opportunity here is to get community-based organizations, activists and organizers and technologists in a room together to really think about how to push back on a lot of these narratives and create counter narratives and forms of data collection.

Justin Hendrix:

So Tamara, I kind of want to put you and Holly in conversation with each other about the harms, specifically. And top of mind, what do we know right now? And of course the AI boom and as Holly, you pointed out, the kind of growth in AI infrastructure is commanding a lot of the headlines at the moment. But can you give us just a little more detail about the state of play in 2025. What we know about how the tech industry interacts with climate change as a phenomenon in particular? And then Alix, I want to come back around to you and ask you a question about what the public knows about these issues, and the asymmetry that we're seeing on some of these issues that I know you uncovered in your report as well.

Holly Alpine:

It's hard to overstate the impact that AI has had in the fossil fuel industry over well, advanced technologies, not just AI. IOT, and these different advanced technologies including AI over the past decade plus. I mean it has helped US oil production surge from 5.1 million barrels a day in 2007 to 13 and a half million barrels today. That's almost a three time fold in US oil production, largely due to technology. I mean, it's extending the shale era that should be phased out, and it is extending the time that we can have an era of shale. And it has transformed companies like Saudi Aramco's operations. They're very, very clear about how this technology is absolutely critical to allowing them to maintain a $3 per barrel extraction cost over the last two decades.

So these emissions are extreme and not accounted for in the tech companies, any sort of carbon accounting. It's not in their 10K, their annual report, their impact summaries about their work with the fossil fuel industry. Basically it's silent, and they're just completely getting away with it, and it is a huge part of their business. We don't have exact numbers on revenue, but we know it is some of Microsoft's top customers are fossil fuel companies. It is not a just marginal part of their business. It is core to Microsoft's business, and that is also terrifying. So that's what I could start with.

Justin Hendrix:

Tamara, I want to give you an opportunity as well. What's the state of play? Can you bring us up to speed on what we know in 2025?

Tamara Kneese:

Sure, sure. So I think as I said before, there's a lot of promises that are attached to AI and also to data centers, in particular. And this notion that when a data center comes to town, it will bring some form of economic prosperity through tax revenue and through job creation. And we hear a lot about the benefits of these infrastructures, and yet if you actually look at the numbers and how things are bearing out, people are literally paying the price for the intensification and acceleration of data center construction. And it's intensifying fossil fuel use, keeping coal plants open longer than they were supposed to be open. And also at the same time, people are paying for that.

And so the people who are on the grid with a data center will notice that their utility bills are going up. And I think the question of land use is also incredibly important. So what does it mean if transmission lines are cutting through a public park? What does it mean if you are giving up really great agricultural land in order to build data centers? What does it mean if you are displacing people from their homes? At the same time, you are also polluting the air around data centers. And so looking at air quality issues, along with noise complaints from people who live around data centers.

And I'll note also that the specter of AI is used as an excuse right now for this kind of investment. But obviously data centers, pre-date generative AI, they're essential for keeping the web going, in general. They're part of cloud computing infrastructure and the entire history of computing is based on systems of exploitation and extraction. It's just capitalism at some point. It isn't completely new and a lot of the harms attached to say the mining of critical minerals and human rights abuses that happen in the DRC, for example. These are not completely new processes.

But what has happened is that this acceleration and hyper investment in AI has led to an intensification of things that were already happening. And so thinking about the harms not just on a sort of environmental justice level, but then also looking at labor exploitation and other forms of harm across the supply chain is incredibly important.

Justin Hendrix:

Alix, I want to come to you and in your recent report, which you've already referenced this case study analysis of five data center developments in Chile, the US, the Netherlands, Mexico and South Africa. One of the things you did is you got on the ground with folks who are in the business of organizing against these things, neighborhoods, communities, activists in these different environments and understood exactly what tactics they were using. But I know one thing I find, and even on the last episode of this podcast, some folks might've heard it, talking to activists in different communities, is this sense of a real asymmetry of information. And so they don't necessarily feel they're armed with the types of fact perhaps that Holly and Tamara are producing all the time. And they can't find independent experts in particular in their local contexts. So I'm just keen to ask you about that, about information asymmetry and what you're seeing on the ground.

Alix Dunn:

Yeah, it's a great question. Just as a quick name check, the authors of this report that I keep mentioning are Prathm Juneja, Hanna Barakat, Emma Prest, and Chris Cameron who each of them took one country and sort of met with a lot of activists who very generously donated their time to talk to us about these dynamics. I think given the rush of interest in this issue area, the civil society folks, the activists that have been working on this for a while are being quite taxed by organizations that want to ask them lots of questions about what they're doing. So thank you to the people that spend time that are helping us put this report together. I just want to sort of name that.

In terms of information asymmetry. I think one of the trickiest dynamics of this is that for a big tech company that is going to roll out a data center in a particular place, it's obviously not the first time they've done that. And so they build up all of this experience making the pitch, convincing local policymakers that all those jobs that they are under pressure to create, they can help with that. Basically all the lobbying efforts, they're just really experienced with it.

Whereas local communities when a data center is going to roll into town, it's the first time they're dealing with it. And I think that sort of fundamental structural aspect of how big tech operates as a sort of multinational, essentially colonial infrastructure that goes in and says, "Hey, we've got this bill of goods to offer you, might you want it?" They are at a dramatic information advantage. And it's not just that they have experience doing this and know the sort of playbook for themselves, it's also that they are kind of shameless. And that that reputational factor where they come in and they've got this big brand name, they've got this big for a local policymaker hearing from a Microsoft or a Meta or a Google or an Amazon, oh my God, this is amazing. This giant company that has all this money is going to help my community.

And that reputation comes into the room with them, which I think is a force that's hard to combat for a local organizer who's like, "Hey, should we ask more questions about how much water this thing's going to take?" So I think in terms of that asymmetry, that is obviously the sort of foundational piece of this that makes it very challenging. I'd also say that companies have gone to great lengths to hide what it is they're doing. So they don't show up as Meta, they don't show up as Microsoft. They show up as some LLC you've never heard of who has money that they're interested in spending to build out physical infrastructure in a particular place. They've got all the experience of those companies and they can name drop them and say, don't you want to be a part of the modern economy, this is going to be wonderful for your community. But they're are some random LLC which changes the dynamics as well.

And Karen Howe actually in her investigative reporting has basically said all work on data centers is by nature investigative journalism now, because they're not disclosing any information about these things. So they're essentially saying, we will withhold as much information as we're allowed to then go into these conversations with local policymakers and communities, with these communities getting caught flatfooted. They've never had to negotiate this type of deal before. And they just come in and they take advantage of that information asymmetry really dramatically. And basically I think hoodwink a lot of communities and convince local policymakers who may be acting in good faith to try and create job opportunities, to try and create growth for communities that are struggling, and say maybe this is the path to that. Even though it often isn't because the jobs created by these are very few and far between post the construction of them because the physical natural demands of these data centers are consistent over their lifecycle.

So as they depreciate, they're still draining water from local communities, they're still draining energy. So I think that has been just part of the challenge of this is that every single community is learning from step zero. I will say there are networks that are emerging that are doing a better and better job of trying to help people who are trying to get up to speed quickly to not just resist, but also just process what it means for one of these things to enter their community. Because some people may decide that this is appropriate for them, but the information asymmetry right now makes it more of a coercion, I think, than any type of democratic process where you would be able to make that decision fairly.

Hanna Barakat, one of the researchers on this paper has been doing work looking at how people in Facebook groups are actually sharing information across communities and there's more and more sort of connected tissue between people that have completed site fights or in them and have learned a bunch of stuff with people that are just entering site fights. And I think that that's a really beautiful thing to see that communities are figuring out ways of being the back office support to be able to do the research and bring the arguments to their local policymakers. Not just a nimby no, but a really informed absolutely not for these reasons. I think the information asymmetry is a core problem, but I think it's changing over time and really quickly as communities realize what's at stake.

Tamara Kneese:

I think that's such an important point, Alix, and I think the case studies are really helpful in this way. So the work that The Maybe has done, the report that Media Justice put out about organizing in the South in particular. And so what's been really interesting, a lot of the work that I'm doing right now focuses on specifically Virginia, which is of course Data Center Alley, the world's capital when it comes to data center construction. And looking at how people there have been working to counter the power of the data center lobby. And so having conversations with folks outside of Virginia who are maybe new to the data center fight, they're asking for materials on how to push back against that the data center lobby, the large tech companies and utility companies in many cases behind them are giving to local politicians. Because they are promising jobs, even though of course data centers don't actually yield that many jobs, especially in the long term.

And the economic benefits are quite murky. Particularly looking at the work that groups like Good Jobs First have done in terms of calculating the subsidies that are going back to large tech companies, which really undermines this idea that municipalities are going to get a lot of money from data center construction. I think what's been really helpful is having these kind of conversations across different sites and providing space for sharing information. And then also thinking about how to make sure that data collected either by communities themselves on the ground or by environmental law folks, or people with more of an organizing background, how do you ensure that that data is also actionable? Because you are fighting such a very strong coalition.

So I think it's been interesting to watch some of the bills that have been in particular places. So in Oregon, they were able to pass a bill that protected ratepayers from paying the literal price for sharing the grid with not just AI data centers, but also chip fabs, thanks to companies like Intel. And I think we've also seen a lot of really interesting policy action and coalitions forming within Virginia. So the Data Center Reform Coalition is another space that we can point to where people are really working together to counter the narratives that are coming from the data center lobby, while also forming their own kinds of data collection to fight that information asymmetry.

Holly Alpine:

I'll say this is also interesting for me to hear because I am a recovering Microsoft data center team employee. I mean, I was on the team that... I was on the data center community development team. Started the program that invested in sustainability projects in those communities. We would go to a new community, I would meet with the local community groups, and talk to them about what their priorities were and how we could invest in local sustainability projects. And I super believed in the projects we did. I think the projects that we did were great. It's just I did become increasingly uncomfortable with the system that I was part of and how that was contributing to an overall, just everything we're talking about. Kind of proliferation of this infrastructure in these communities who didn't fully understand or we weren't giving the full information on the scope and scale of what that rollout would look like. Yeah, so it's so interesting to come from that world to this and I'm still thinking holistically just how I even think about it. So yeah, thanks for the conversation.

Justin Hendrix:

We have the benefit, I should say for any of our listeners, most of the time the Tech Policy Press Podcast is recorded on a lonely Zoom, but we have of course live listeners in this context. And so I want to refer to some of their questions in the Q&A. And so one of the first questions is really around the kind of commitments that companies in the tech sector have made around trying to be less energy intensive in their work. Or I might extend the question to maybe other environmental commitments that the companies are making. Maybe I'll just ask, and I think I know the answer based on your prior comments, but can each of you comment on how far off we are on the kind of commitments and claims that the companies are making about their environmental practices and where things really are?

And I want to recognize that there's a difference sometimes between the kind of corporate PR and the sort of things that they say, maybe at an event like Climate Week when they come to town, if in fact they do. But maybe their earnings reports and their other kind of investor statements are a bit different. What do they claim they're doing on these issues versus what they appear to be actually doing? Does anybody have a comment on that? Go ahead, Holly.

Holly Alpine:

So Microsoft has commitments around carbon water waste and ecosystems. The carbon commitments that was set back in 2019 I believe was around getting to carbon negative, and that includes their scope one, two, and three. So basically in their supply chain. So I actually think Tamara and Alix might have more information on how or if or how they're slipping on those commitments. My point is that Microsoft's technology that is being used to have a five to 10% increase in recovery of existing oil wells across large fields, adding up to millions of extra barrels annually is not included in any of these tech companies' reports.

It's not part of the current GHG protocols. And so we don't know what the total emissions are, and right now there is zero responsibility for it. We can't really say how they're meeting their commitments because there are no commitments. There is one thing that Microsoft will point to if you ask about this issue, which has happened many times. They've worked with a lot of journalists who have asked Microsoft to comment, they will point to their energy principles, which state they'll only work with oil and gas companies that have a net zero commitment.

Which at first sounds great until you realize that it's a 2050 commitment, that's way too far away. There are no interim targets. And the biggest part of it is there's no standard that they have to follow. So generally, these companies that are setting net zero targets are only talking about their own operations, how they're powering their office buildings or driving an electric car around the oil rig. It has no commitments at all. There's no accounting for the fuels.

So that's something that we are doing right now. The Enabled Emissions Campaign, the first thing we're trying to do is an energy economic analysis and a quantification analysis of how big is this issue. And some of our initial quantifications that we've seen is quite large, so we're doing that right now. And also quantifying how AI alters fossil fuel supply economics, how it lowers costs and boosts productivity and optimizes logistics. So I'm getting a little off topic here on Microsoft's commitments, but basically, long story short, it's not part of any of these tech company's commitments and we think that it should be, or at least a principled approach, and that's one of the things we're advocating for because we don't know how big it is and they're not going to tell us unless they're forced to or we do the quantification ourselves.

Tamara Kneese:

And I'll just jump on and say that it's been really fascinating to watch how Net Zero has really rescinded, has been rescinded. So a lot of companies were clamoring to make net zero pledges and were competing with each other over their commitments. And in some cases there were employees within companies that were really driving some of that change internally and pressuring companies to make stronger climate commitments and to adhere to net zero goals. But now we have companies saying outright that Net Zero itself was a moonshot. And so this idea, well, we were never really going to be able to do it, and yet we're supposed to believe that nuclear power is going to save us all and provide us with endlessly abundant energy.

But I think what's really interesting in this moment is how Net Zero, which had been in many ways part of the green-washing on the part of tech companies, is now out of fashion and has become almost a liability. And so as we see net zero, and any kind of sustainability commitment become further politicized and thinking about how now the conflation of diversity, equity and inclusion or DEI, and sustainability, and we see this again and again, especially with how the current administration is thinking about federal funding and threatening companies that have any sort of DEI or climate related commitments. I think it really gives companies an out to just say, throw their hands up and say, look, AI is really important, and if we just invest enough in AI and invest enough in speculative forms of power, then maybe we'll be able to solve climate change.

Justin Hendrix:

I actually do want to ask you specifically about that narrative, and there's one question that we have here from a listener that I think kind of gets to this a little bit. The sort of claims that some of the investments, certainly in artificial intelligence, are ultimately going to be the solution to consumption, or the solution to our climate challenges. And we're hearing that case made more explicitly, it seems like by tech leaders who are basically saying on some level, let's burn all the fuels now. Burn all the fossil fuels, now dig it out of the ground, burn it, let's convert it into energy and convert that energy into artificial intelligence. Because ultimately that's the only path forward here to solve the climate change issue.

And somewhere on the spectrum of, ‘we'd like to build data centers that are more sustainable and use more renewable energy,’ through to, ‘hey, burn all the coal now as fast as possible so we can get to AGI as fast as we can.’ I don't know, I mean, what do you make of those types of claims, Alix? We're hearing some of that from Silicon Valley, almost a kind of sort of like a suicide pact almost. Let's burn it all as fast as we can to get to that point.

Alix Dunn:

Yeah, I mean I think it's a sickness, which I think lets them off the hook a little bit. Because I think it is also an incredibly cynical set of arguments that they're making because they know that they're in a pretty precarious financial spot where they've made tremendous amounts of promises that aren't converting into revenue. And now they're like, how can we embed ourselves so intensely into infrastructure? So sell our services into government, become government infrastructure in a way that government can't undo. Be so embedded in every aspect of our lives that you can't undo it. And the way I've been thinking about it is that they know they're going to need a bailout.

And so it's this race to become integrated meaningfully enough into all of our infrastructure that we can't just say, you guys made a huge bet. People were telling you this entire time that it was not going to pay off and now you failed as businesses. They want to put us in a position where we can't actually do that, where we're going to have to negotiate with them because their services are essential. So I think on the longer arc, I feel like it's coming from a place of fear, this kind of infinity industrial complex. Like energy is infinite. But then you see actually the calculation of that with, for example, Meta's data center in Louisiana that they're now going to have to bring on, I think three new natural gas plants to be able to power that center.

So they're already sort of admitting that they're outside of the bounds of the renewable grid infrastructure that we've been building as fast as we can, but it's not fast enough to accommodate an industry that is churning out energy at this rate. And so it just feels like a really elite, irresponsible use of a dwindling amount of resources and a smaller and smaller window of action for us to be able to take to convert our society into something that is sustainable. And I feel like it's also sucking up a lot of oxygen in the room. So that idea that we should just burn all everything and convert it into energy, convert it into compute so that we can somehow, I don't know, what happens then? I don't actually know what their argument is.

Like the Eric Schmidt argument, it's just really cynical. He's going to buy himself 15 more years. He's then going to retire, die in Cyprus or wherever he lives with his 17th house and not have to ever deal with the chaos he's wrought on the world. And I think that we shouldn't let them get away with it. We should challenge these narratives wherever they happen. And we should also force them to face reality and facts about what these facilities are taking out of local communities. I think going back to Karen, how reporting in Arizona that the local policymakers when FOIA-ed, so Karen used the Freedom of Information Act to say, show me what the water demands are projected to be in this new data center as part of their zoning application.

They redacted those figures saying that those figures are commercial. It's a commercial secrets like information that is private and needs to be kept private to protect the interests of business. The idea that you would protect the information about something like that, rather than try and protect your community's water. I feel like we're just seeing those kind of situations constantly where it's like they don't want to put all the information on the table and then have a very reasoned conversation about is this worth it? Because if we have that conversation, they know what the answer's going to be and I think they'd be in a lot of trouble at that point.

So to answer the question in the chat, no, I don't think it's worth it. No, I don't think we're having good concrete conversations about the trade-offs because they don't want us to have that conversation. They want us to be so far gone into committing to their vision that we can't go back.

Justin Hendrix:

I want to ask a question that is similar to one that's been put forward by a listener here about where you all see gaps in efforts to address some of these harms, or to build climate justice groups or other social movements that might resist some of the types of harms that you are describing here. Each of you is involved in talking to and observing, and in some cases assisting communities that are working on these issues. Where do you see the gaps at the moment?

Tamara Kneese:

It's a great question. So I think part of the problem is that the forms of data that people on the ground need, particularly around public health impacts, tend to be site specific and very situated. And so there have been a few really great general overviews of the public health impacts of data centers. And this builds on previous work on say the public health impacts and air quality issues around warehouses, for example, and other forms of industrial pollution. But the issue is that a lot of the information that communities really need very quickly in order to engage in the site fight is not readily available.

And so getting permitting data, really understanding when a data center is being permitted. So I can give an example of a data center complex that was being planned in San Jose closer to where I live. And so in California, we don't have the kind of massive hyperscalers that we see growing in places like the US South because we have environmental protections in the state of California, at least for now. And there are places that are actually portrayed as a, quote-unquote, green or a sustainable data center. And so yes, a data center might be using renewable energy for the most part, but they may still be using diesel backup generators.

And in one case, there was a homeless encampment that was being cleared in San Jose to make way for the construction of a data center. Now, there are a ton of activists on the ground in San Jose, and this is a very activist oriented place, the Bay Area, but there was not enough time for people to organize around it. And I think that's part of the problem. If you don't have a very long period of public comment, if people are not aware of town hall meetings that they should show up to, that can make it much harder. But talking to folks in Virginia, like Savannah Wilson at Clean Virginia has made it very clear that sometimes you just need 15 or 20 people to show up to the meeting to show local officials that you care. And that creates the leverage for negotiation that people need.

And so I think there's not only a problem of information asymmetry and the problem of timing, and that is certainly a gap, but then also the issue of transmission lines. So thinking about the fact that these often will cut across state lines. And so people in one state, you might know that a data center is being planned in Virginia, actually the transmission line is going through Maryland or West Virginia. And so the people who are also going to be affected don't necessarily have any awareness of what is being planned. And I think the sheer number of sites right now that are being planned is also overwhelming. It feels like whack-a-mole.

And I think part of the problem too with that can be a problem with just stopping a data center in one place is that it's very likely that it will pop up in another. And so finding ways to coordinate is also going to be increasingly important.

Justin Hendrix:

Holly?

Holly Alpine:

Thanks for the question. There are a lot of gaps, huge gaps. I mean, despite this scale of this threat and advances in AI and other technologies that are reshaping the economics of fossil fuel production and keeping oil competitive and slowing dramatically the transition to renewables globally. It's a huge blind spot. There exists no frameworks or governance for how the tech can be used. It's absent from any carbon accounting or AI policies or corporate safeguards.

We've seen that corporate voluntary measures that exist are wholly insufficient. I think one of the reasons for this though is there's a lot, but it's big and it's intractable. It's difficult, it's system level, which you don't have the sites that you can go and see exactly how these people are being impacted. We don't have that in this context. So that makes it challenging, but it doesn't mean we shouldn't try and do something about it.

And we think this regulatory gap really is a clear opening to draw on proven models and build guardrails on AI systems that materially enable fossil fuel expansion. For example, and maybe I'm answering another question here of what can actually be done. The EU AI Act that has come out, it prohibits certain high-risk applications of AI, such as social scoring or biometric surveillance based on societal harms. And we don't think it's a big leap or even a leap at all to say that's climate change at massive fossil fuel investment and expanded production is a societal harm.

So we think that this application should be classified as high-risk and various other measures we can get into. But to get back to your question on gaps, big regulation and guardrail gaps, there really exist no guardrails today. And that is a big part of our work is trying to propose what those guardrails could look like.

Alix Dunn:

I'll just add, I mean I think one of the gaps I see is that people that have technical expertise and are interested in the technology dimension of long-standing social and political challenges in the world, oftentimes like Columbus movement organizing and say, ah, this is an issue I care about. I'm now going to start from ground zero of figuring out how to organize my community against this. And I don't think that period lasts for very long because you quite quickly realize that there are movements that have been existing and in operation and in solidarity with each other for a really long time.

And so I would say one of the gaps is that the people that come with this understanding of what's happening with the sort of AI infrastructure buildup to seek the people that have been doing the work within their communities for a long time, one of the things we found in our report is that there are many movements that have sort of taken data centers as part of their history of resistance against incursion, against their access to water rights, for example. There are communities that have been working on this for a long time, and I think if you can speed the process up, if you just seek those people rather than start doing work in this area, then discover that they've been doing this work.

So I would say part of it is that kind of mix between social justice and political organizing and tech policy people that there needs to be just more intentional effort at figuring out who's got strengths where, how might these movements connect to each other to have an impact. And then I would also say that I feel like this whack-a-mole point that Tamara makes, I think is a really important one. And I think it means that we need to be much more strategic in thinking about rights-based approaches to these questions and things that might give us broader based wins that allow us to protect, regardless of where you are.

And I think sometimes we settle in on the very transactional local fights and forget that there are frameworks of law and policy at the EU AI Act as an example that Holly just mentioned, that I think we need to think more creatively about and be much more radical in what we're asking for. Because I think right now we've been very "mother may I" about setting rules on these things because we don't want to seem naive, I think, in terms of where the economy might be going and seeing the people that are like, no, please progress don't happen. You don't want to be that person. But I think we then miss opportunities because I think there are a lot of people in the general public that want to see more radical protection of their water, of their air, of their communities. And I think that we would get a lot of support if we were to think more structurally about the rights we should be trying to protect for people.

Justin Hendrix:

I want to just give you each opportunity to say a last word. But also if you will, for the listener, one thing they could do to get involved in helping set the climate agenda for tech. What would you recommend individuals that are listening to this go and do? It could be engaging with your organization somehow. It could be something else you'd like to see individuals go and do. Alix, I'll start with you.

Alix Dunn:

Yeah, I mean, a very small plug specific thing. Next week we are really pleased to host the organizers who have been doing resistance work in Arizona, Louisiana, and Tennessee in New York at Climate Week. And they're going to be sharing the work they've been doing and hopefully be sort of setting the story straight within the Climate Week agenda. Because a lot of times those spaces can be very elite, very national, very abstract, and we're really hoping that those voices will contribute and reshape some of the Climate Week conversation. I can share the live stream link for those that want to join that, but I think it'll be an incredible conversation. They're just amazing organizers who've been doing incredible work. So that's one thing.

And then I just think broadly, I feel like we need to start getting specific with what we're talking about and be more comfortable pressuring these companies to answer the question. Whenever they say this is going to benefit humanity, we need to be asking more aggressive questions of what are you talking about? What is the benefit? Show it to me, show it to me. Force them to actually answer that question. I think once they're forced to start actually engaging in that question, the trade-offs, the dramatic trade-offs we're being asked to make as a society will very quickly become obviously silly and non-starters.

But I think until we start putting the feet to the fire of these narratives and of these people and these leaders, I think we're fighting an uphill battle.

Tamara Kneese:

So yeah, I'll be at Climate Week next week. We have a few events that Data & Society is putting on or contributing to. But I'd say that maybe the one that is most exciting to me right now is the event that we're doing in partnership with the Natural Resource Governance Institute, which will bring together people who are thinking about AI, and also people who are thinking about human rights and critical minerals and natural resources in general. And so I think that conversation will be quite interesting. So plug that, I guess that'll be happening on Thursday. Yes, at the International Labor Organization Office.

And yeah, I would say that Alix's point about thinking about histories of organizing and the need to situate these struggles as well. So I'll just also plug the series that I put together for Data & Society, which is the Cloud is Dead. Looking at histories of living with the legacies of resource extraction and hearing from people who are really thinking about the deeper colonial and historical relationships between particular places and technology, and also looking at histories of resistance.

Justin Hendrix:

And Holly, I'll leave it to you for the last word.

Holly Alpine:

Great. Well, folks can always go to EnabledEmmissions.com, sign up for our newsletter if you want to tell us about what you're interested in or how you want to contribute. If you have certain skills and you want to get involved, just include that when you sign up. I read all of those. I reply to a lot of them. I'm very organized, so I will log it away and reach out to you when needed. But building on what Alix said about forcing companies to answer how their tech will be helping the world, also force them to answer questions on how they are stopping the bad. An analogy here is when you're getting healthy, you can start eating carrots. That's a good thing. But if you're also still smoking a pack a day, and eating a dozen donuts per day, you need to stop doing that to get healthy. And I think it's a good analogy here.

I'll throw a last analogy around bringing in what the technology is used for into the environmental harms of technology discussion. Which is, I think another thing everyone can do is when we talk about environmental harms of the technology, just be sure that we're also including what that tech is used for. It's like we were building a robot and you're looking at the environmental footprint of the robot and focusing on how it's built and the power it's running on. That's super important. Also, if the robot is going out and mowing down the Amazon and dumping gasoline into rivers, we need to look at what that robot is actually doing as well. And if it hypothetically could also care for the elderly or pick up trash, that's great, but that doesn't mean we should also keep allowing it to destroy the rainforest.

And that's kind of what the conversation is centered on right now. It seems like we're starting to finally look at how that robot is built, what it's running on, and then focusing on hypothetically the good things it could be doing and almost entirely leaving out of the conversation what it is currently doing and has been doing for the last decade. So bringing that into the conversation. I'm so excited for Climate Week and all these great conversations and hope to bring that in as part of what we're talking about and hopefully building different frameworks so we can have this robot be a force for good. That's my spiel.

Justin Hendrix:

Holly, thank you for leaving us with that visual. I think it's a useful visual and metaphor and a good place for us to end. I want to thank you, Holly, Alix, Tamara, thank you so much for taking the time to speak to me today. And also to everyone who dialed in from around the world. I know we had people from Michigan to Kenya to India and beyond I know that introduced themselves there. So grateful for all of you and hope some of the resources that you've encountered here are useful. So with that, I'll thank the three of you and wish you a good day.

Holly Alpine:

Thank you so much.

Alix Dunn:

See you soon.

Tamara Kneese:

Thank you.