You’re out of free articles.
Log in
To continue reading, log in to your account.
Create a Free Account
To unlock more free articles, please create a free account.
Sign In or Create an Account.
By continuing, you agree to the Terms of Service and acknowledge our Privacy Policy
Welcome to Heatmap
Thank you for registering with Heatmap. Climate change is one of the greatest challenges of our lives, a force reshaping our economy, our politics, and our culture. We hope to be your trusted, friendly, and insightful guide to that transformation. Please enjoy your free articles. You can check your profile here .
subscribe to get Unlimited access
Offer for a Heatmap News Unlimited Access subscription; please note that your subscription will renew automatically unless you cancel prior to renewal. Cancellation takes effect at the end of your current billing period. We will let you know in advance of any price changes. Taxes may apply. Offer terms are subject to change.
Subscribe to get unlimited Access
Hey, you are out of free articles but you are only a few clicks away from full access. Subscribe below and take advantage of our introductory offer.
subscribe to get Unlimited access
Offer for a Heatmap News Unlimited Access subscription; please note that your subscription will renew automatically unless you cancel prior to renewal. Cancellation takes effect at the end of your current billing period. We will let you know in advance of any price changes. Taxes may apply. Offer terms are subject to change.
Create Your Account
Please Enter Your Password
Forgot your password?
Please enter the email address you use for your account so we can send you a link to reset your password:
Geopolitics, the heightened importance of climate change, and the sheer size of the conference have transformed the event into something that it was never meant to be.

It didn’t attract a lot of attention, but for a few months, it looked like the United Nations climate process might break down.
There, process is substance: One of the most important acts every year is the selection of the next country to run the Conference of the Parties to the United Nations Framework Convention on Climate Change, or COP. This distinction normally rotates among the UN’s five regional country groups; next year, a country in the “Eastern Europe” group is due to host. All the members of a group must unanimously agree on which country will get to host.
This is a highly contingent way to decide who gets to host a climate conference. Really, the entire schema of UN regional groups represents a hangover of Cold War geopolitics that is now indefinitely unchangeable. (The “Western Europe” group is essentially the early members of NATO; it includes such notably non-western-European countries as Turkey, the United States, and — hilariously — Australia.)
The “Eastern Europe” group, meanwhile, amounts to more or less the former members of the Warsaw Pact. For obvious reasons, these countries cannot agree on a consensus choice in 2023. Russia, the group’s largest member, was not amenable to holding the COP in any eastern Europe NATO member state, such as Poland, Latvia, or Finland. The eastern European NATO members — as well as Ukraine, which is also in the UN regional group — were similarly opposed to holding the COP in Russia.
That meant that attention focused on the group’s countries in the Caucasus, at the edge of central Asia: Georgia, Azerbaijan, and Armenia. Yet difficulties presented themselves here too. Azerbaijan successfully seized an Armenian exclave earlier this year, evicting up to 120,000 Armenians as part of a campaign described as ethnic cleansing. Armenia blocked any Azeri bid to host the COP.
For the first time in the UN Framework Convention on Climate Change’s history, no country would have been able to lead COP the following year. Geopolitics had seemingly broken the consensus mechanism that makes the climate conference work.
This amounted to more than just a deficiency in party planning. It would have forced Bonn, Germany — the home of the UNFCC’s permanent headquarters — to host COP29, a kind of “break in case of emergency” default option. And it would have allowed the United Arab Emirates, a petrostate that has reportedly used the COP to make oil deals, to retain the conference presidency for at least another year.
That didn’t happen. Late last week, Armenia lifted its block of Azerbaijan’s bid, and the two countries mutually released prisoners in a gesture of good will. (Their rapprochement happened suspiciously close to President Vladimir Putin’s visit to the U.A.E.) Next year’s COP will seemingly happen in Baku, the Azeri capital.
But just because the COP process didn’t break doesn’t mean that it’s not being stretched. All is not well with the COP. During this year’s conference, a picture emerged of a COP being tested by a more rivalrous, conflict-prone world. Geopolitics, the heightened importance of climate change, and the sheer size of the conference have transformed the event into something that it was never meant to be.
This year, more than 100,000 people attended the COP. It was held at Dubai’s opulent Expo City, the Disney World-style convention campus initially built for the 2020 World Expo, the modern successor to World’s Fairs. Hundreds of nonprofit groups and companies, as well as more than 190 countries, ran public pavilions that advertised their climate accomplishments and views on decarbonization. Negotiators divided into different blocs: China and the United States, oil-producing states and small island nations, the West and the rest.
It wasn’t always like this. When the first COP was held in Berlin in 1995, the world was in a very different era, Lee Beck, the senior director for Europe and the Middle East at the Clean Air Task Force, told me. It was “the peak of multilateralism, followed by relative geopolitical stability and peace,” she said. The United States and the broader West set the agenda for global events.
“In the last two years — others would say the writing was on the wall as early as 2014 — geopolitical fragmentation really is visible,” she said. “You’re really pushing the limits of multilateralism at this one. One of the cracks is we’re unable to agree where the COP even will be.”
But geopolitics are not the only force stretching COP to the limit. Another is the sheer size of the event itself.
There used to be “big COPs” and “small COPs”: COP21, the 2015 meeting where the Paris Agreement was finalized, was a “big COP,” but the following year’s conference in Marrakech, Morocco, was a fairly minor one. Even COP21 was less than half the size of this year’s COP. And in one possible read, this year should have been a smaller COP — the biggest to-dos were formally launching the Loss and Damage fund and writing the Global Stocktake report, a kind of report card on the world’s climate progress (or lack thereof).
But small COPs don’t seem to happen any more. Since the pandemic ended and COP26 took place in Glasgow, Scotland, COPs have swollen in size, creating the age of the new “mega-COP.” More than 100,000 people attended the conference this year, making it by far the biggest COP ever. It was more than twice the size of last year’s confab in Sharm al-Sheikh, Egypt, which was previously the biggest COP ever. Most of those attendees had nothing to do with the negotiations ostensibly at the center of the conference — they were investors, technologists, scholars, scientists, or experts — and instead made up a de facto global trade show on climate solutions.
COP is now so big and climate is now so important that even the lack of news about the conference can generate news. When President Joe Biden declined to attend this year’s conference, The New York Times push-alerted it.
But there are possibilities that could improve the situation. One of them might be that COP simply becomes so unmanageable that it has to scale back. Few cities have the spare capacity to house an extra 100,000 visitors for 12 days. New York City, for instance, only has about 123,000 hotel rooms total. If COP were to keep growing, the problem would only get harder. When 150,000 people descended on San Francisco for Salesforce’s annual conference in 2015, the company docked a cruise ship in the bay to provide an extra thousand rooms.
There are solutions, Beck said. She noted this was the first year that every continent had held its own Climate Week: a smaller event focusing on more region-specific decarbonization challenges. This COP has also seen the emergence of country coalitions that rally around different issues or approaches. The set of countries that backed a pledge to triple renewable capacity, for instance, is different from the smaller coalition that wanted to triple nuclear capacity. These smaller, more sector-specific coalitions may have more ability to actually decarbonize and address climate change, she said.
For all these challenges, perhaps the biggest miracle is that the UNFCC process works at all, Eve Tamme, a former climate negotiator for the European Commission, told me.
“The UNFCCC process is based on consensus between almost 200 countries. Judging based on the complexity of the issue at hand and the divergence of views among the countries, it seems impossible that such a process could deliver anything at all,” she said. Even when you follow the negotiations closely, it may seem like there’s barely any movement at all, she said.
“But then again, we got the Kyoto Protocol,” she said. “And we got the Paris Agreement. So while it may look broken in the short term, somehow this dysfunctional process can still deliver.”
Log in
To continue reading, log in to your account.
Create a Free Account
To unlock more free articles, please create a free account.
Building a data center is also quite carbon-intensive.
When I helped start Heatmap News three years ago, I didn’t think I would be writing this much about big tech companies.
I knew that, sure, they were crucial to America’s ability to develop and scale some next-generation emissions-reducing technologies. (By then, Microsoft had already started its huge carbon removal purchasing program.) And, yes, I knew they bought a lot of renewables. But I still understood their clean energy programs chiefly as an employee perk — a way for some of the economy’s richest firms to show their largely urban, college-educated, and liberal employees that they cared.
Perhaps that was true once. It’s not true anymore. Over the past several years, the tech companies have become major electricity consumers and producers in their own right. Artificial intelligence has turned their electricity procurement and development businesses into core operational competencies. (Meta and Microsoft have even considered entering the electricity trading business.) Some of the thorniest questions in climate policy were first encountered by these tech companies.
More importantly, their hunger for electricity has transformed them into quasi-industrial companies — and given them enough heft in the market to sometimes counterbalance (and sometimes collaborate with) the utilities and fossil fuel firms that previously steered the sector. As such, they’re now crucial parts of the U.S. decarbonization story.
Three companies in particular dominate the artificial intelligence cloud business: Google, Amazon, and Microsoft.
The country’s best-known frontier labs, such as OpenAI and Anthropic, rely on these companies to provide their compute power; Amazon Web Services is the backbone of virtually the entire online software industry. Amazon, Google, and Microsoft account for more than half of the country’s data center power capacity, according to the investment firm Jeffries.
So these companies’ emissions are, in a sense, not only their own; they also give us a view into the AI industry’s carbon footprint more broadly.
Over the past two weeks, all three of these cloud providers released their energy and emissions data for the past year, and we’ve looked at the top line findings from these reports in past editions. Today I want to briefly dive into what they could mean together.
Let’s handle the part you already know: Everyone’s emissions are up.
Microsoft’s emissions grew by 25% last year, their largest year-over-year leap since the pandemic. Amazon’s emissions leapt by 16%, its largest one-year increase ever. Google’s emissions increased by 18%, rising above their pre-pandemic level.
This surge will make the companies’ climate goals increasingly difficult to meet — and some of them are coming up fast. Microsoft has pledged to become ‘carbon negative’ by 2030, meaning it must remove more climate pollution from the atmosphere than it emits in that year. Google has pledged to achieve net zero by 2030, a goal that requires — by its own estimate — cutting its emissions in half by that year, as compared to their 2019 level. Amazon, meanwhile, has pledged to achieve net-zero in its operations by 2040.
All three firms’ greenhouse gas emissions are up because of the AI data center boom. Microsoft consumes nearly four times as much electricity as it did before the pandemic; Google’s electricity use has more than doubled.
These companies’ energy use has swelled, too, but at least as of last year, nearly all of their energy demand still took the form of electricity. When we think about “electrification” in the national context, perhaps we should think at least as much about these AI megalodons as we do about heat pump or battery manufacturers.
Amazon, to its shame, does not publish recent electricity usage data, so it doesn’t appear on either of these charts.
But outsiders have estimated its power consumption based on the numbers it does publish. Hendrik Rood, an IT researcher and consultant in the Netherlands, calculates that Amazon’s data center business used 78,000 gigawatt-hours in 2025. That would mean it consumes nearly as much electricity as Microsoft and Google combined.
As I cautioned yesterday, some of these figures are already outdated. Although all three companies just released their 2025 sustainability data, Microsoft brackets its report to the fiscal year, which ended on June 30, 2025. Google and Amazon’s data covers the calendar year.
In what might be a quirk inherent to the genre, all three sustainability reports have a somewhat defensive tone (or at least a writing style that tries to anticipate quibbles). These companies know that their sustainability pledges, embraced in the heady flush of 2020 and 2021, have become much more difficult to fulfill in the AI era. And they want you to know that all of their emissions could be worse — if not for their corporate policies, pollution might be much higher.
I can’t say I find these counterfactuals entirely believable. We don’t know what Google or Microsoft or Amazon would do if, say, computing were more energy intensive or a certain process more environmentally damaging. And Jevon’s paradox suggests that every gain in efficiency — especially for a service as in-demand as AI — will make it cheaper to use AI, therefore raising its energy demand.
But I do think it’s worth sharing these claims to get some perspective. Google, for its part, says that its corporate emissions would be five times higher than they are if not for its total slate of policies:

Microsoft takes a more clinical approach. It selects four of its corporate policies: “carbon-free electricity, sustainable fuels, XBOX console efficiency,” as well as efforts to decarbonize its Surface tablet production. If not for these interventions, it says, it would have emitted 34 million tons of greenhouse gas into the atmosphere last year, not the 21 million tons that it did produce.
For all the focus on the difficulty of powering data centers (including by Heatmap), electricity does not drive most of these companies’ emissions — or it didn’t in the first half of last year, at least. The majority of Microsoft, Google, and Amazon’s greenhouse gas emissions came from what are dubbed “scope 3” emissions, a somewhat nebulous category that includes buildings, employee travel, and the full carbon footprint of their supply chain. This category reflects the AI boom in its own way.
(Skip this if you’re a sustainability nerd: In the classic schema used for corporate emissions accounting, “scope 1” emissions are direct fossil fuel pollution from an asset that the company owns or controls, “scope 2” emissions are pollution associated with the electricity, steam, or chilled water purchased by the company, and “scope 3” emissions are everything else — pollution from the company’s upstream supply chain and its downstream product use. I find this scheme makes somewhat more sense for businesses like airlines and automakers than it does for technology conglomerates. But that’s a different newsletter.)
It makes sense, then, that Amazon should have huge scope 3 emissions. The scope 3 subcategory called “Purchased Goods and Services” drives the largest share of its emissions; these include pollution from goods and services that Amazon buys for its employees to use, as well as all the embodied carbon in its line of Amazon Basics products.
But the biggest driver of scope 3 emissions — and thus for emissions overall — for Microsoft and Google came from “capital goods,” a category that covers new construction, physical assets and other fixed infrastructure used to produce products and services. More than 40% of Microsoft’s total emissions came from capital goods, and they made up more than 9 million metric tons of the company’s greenhouse gases. Google doesn’t fully aggregate out its “capital goods” category, combining it with the “use of sold products” subcategory, but it was responsible for almost 9 million tons as well.
These capital goods include the new data centers themselves: all the cement, steel, server racks, and silicon that actually make up the physical infrastructure supporting the AI boom. Here at Heatmap, we often focus on the electricity sector because it’s where so much change. But it’s good to remember that construction remains enormously carbon-intensive, and the literal buildings that house AI are, in many cases, still driving a disproportionate amount of emissions.
The July 4 heat wave showed just how far the metropolis has to go to reach its decarbonization goals.
New York City’s decarbonization plan has stalled. The events of this year’s Fourth of July weekend all but prove it.
The temperature in the city reached as high as 100 degrees Fahrenheit on Thursday, July 2, the hottest it’s been here in 14 years. As New Yorkers blasted their air conditioners to stay cool, utilities drew on all of New York’s resources to serve the resulting electricity demand for cooling. These included a fleet of dual-fuel power plants, which can burn both oil and natural gas and encompasses many of its peakers, which turn on to deal with spikes of demand.
Those dual-fuel plants pushed over 10 gigawatts of electricity onto the grid on the evening of July 1— about a third of the total load in the state — and hit similar peaks on the 2nd and 3rd. The peaker fleet owned and operated by the New York Power Authority was operational for over two-thirds of the heat wave, which persisted for four consecutive days, while some ran nonstop from 7 a.m. July 2 to 3 a.m. July 4, according to NYPA.
In response to questions about the use of its peakers during the heat wave, a NYPA spokesperson told me, “During times of peak energy demand, like last week’s heat wave, the state’s independent grid operator called upon NYPA’s Small Natural Gas Power Plants to run well beyond their typical usage to meet high energy needs and prevent localized blackouts.”
While specific generator information is a protected trade secret, they said, “capacity suppliers are critical resources to meet system peak loads like those experienced during the recent heatwave.”
And yet still, over 100,000 people lost power during the heat wave. Real-time electricity prices in the area of the New York grid that includes the city got as high as $1,465 per megawatt-hour on the evening of July 3, according to data collected by Grid Status.
At the same time, the latest addition to New York’s non-carbon electricity generation fleet, a transmission line from Quebec that can transmit up to 1,250 megawatts known as the Champlain Hudson Power Express, was struggling. It experienced an unplanned outage on July 1, the first day of the heat wave, followed by a second outage beginning on July 4 that still had not been resolved as of Friday.
Since 2014, the city has had an aspirational goal of reducing emissions by 80% of its 2005 levels by 2050. CHPE was a major part of that plan, which also included offshore wind and utility-scale solar. There has been progress: Of the 1,000 megawatts of solar the city aims to have installed by 2030, about two thirds have been built. Even so, about 90% of New York City’s electricity came from fossil fuels in 2025, according to the city’s comptroller.
Why the difficulty decarbonizing? Blame a mixture of policy and geography. New York City is dense and has a lot of old buildings with old heating systems. Reducing consumption of fossil fuels requires getting cars off the road (congestion pricing) and retrofitting buildings with electric appliances (Local Law 97).
But that’s the demand side — the supply side is far trickier. Utility-scale non-carbon-emitting power on the orders of hundreds of megawatts or a gigawatt will have to be built elsewhere and piped in via transmission lines. That means offshore wind, solar (ideally with battery storage), and maybe one day nuclear power.
To the extent New York City can build solar and storage locally, it means dealing with a thicket of building regulations and local opposition. Efforts to shut down or replace peaker plants in the city have run into a brick wall at the New York Independent System Operator, which has declared that at least some peakers will have to stay online through the end of the decade to maintain system-wide reliability.
The only other new source of carbon-free power currently under construction is the offshore wind project Empire Wind, due to come online in 2027. NYISO said last year that without CHPE, Empire, and two local transmission projects planned to enter service by 2030, New York City would be “deficient in the summer” through 2030.
Of course developers have scrapped several other offshore wind projects over the years, whether due to problems procuring the right size turbine or the Trump administration buying out their lease. And though New York Governor Kathy Hochul pledged last summer to develop at least a gigawatt of new nuclear capacity in the northern region of the state, that is probably a decade away from fruition.
Meanwhile the Clean Path transmission line, which was meant to connect New York City to several gigawatts of new wind, solar and hydropower, saw its contracts canceled in late 2024 as its projected costs continued to rise. Last year, utility regulators shut down an effort by the state-run New York Power Authority to take it over as a “priority transmission project,” questioning whether it was “needed expeditiously” to meet downstate reliability needs and arguing that the project “will not be needed to serve substantial amounts of generation until well after 2033, and possibly not until 2040.”
While the city has some utility-scale battery storage systems, would-be developers have faced intense local opposition. Fullmark Energy, for instance, scrapped a planned 650-megawatt storage project after protests from political figures, including frequent mayoral candidate Curtis Sliwa. A dispute over another battery storage project in Queens has escalated into accusations of assault leveled by Councilmember Phil Wong, who called for a criminal investigation into what he said was an assault by a contractor for a project against his staffer.
So what’s left for New York City to do?
Any near-term progress will likely come from increasing efficiency and adding marginal generation capacity, as opposed to large-scale new projects and decommissioning of power plants.
“We need to max out our energy efficiency gains from Local Law 97,” former New York City Chief Climate Policy Advisor Daniel Zarelli told me, referring to a 2019 law mandating steep reductions in emissions from large buildings in the city, which came into effect two years ago. He also called for a further“push on batteries and behind the meter solar, clean energy, and energy efficiency.”
Already across the state, behind-the-meter solar is shaving off peak power demand. On the afternoon of July 2, behind-the-meter solar accounted served about 4.5 gigawatts to users, according to NYISO and Grid Status data.
Going forward, Zarelli said, the city should use its purchasing and planning power — as it did with CHPE — for projects like resurrecting Clean Path. “We need to be starting now. Maybe it’s not by 2030, but soon after we could be getting the benefit of that.”
“Battery developers started to see interconnection costs that were around 30 or 40 times what is standard,” Patrick Robbins, director of the Utility Customers Association told me. “It just means that new battery projects completely don’t pencil out and so we have a de facto moratorium on new [battery] projects.”
Advocates for solar and storage have blamed Con Edison for the city’s slow progress there, claiming that changes in the interconnection process have made it essentially cost prohibitive for battery storage developers to move forward on new projects.
Some of these fights have landed in front of New York’s Public Service Commission. In a filing, the city cited data from Con Edison showing that “the interconnection costs for some projects … have increased by thousands of percent,” citing one project whose interconnection costs jumped from $640,000 to over $35 million due to changes in how Con Edison attributed grid costs from new projects.
"Battery storage is essential to New York's clean energy future, and Con Edison strongly supports the development of energy storage when projects are deployed at the right locations, at the appropriate scale, and with operating parameters that provide the greatest benefit to customers and the electric grid,” a Con Edison spokesperson told me. “Because grid constraints vary across our system — from neighborhood‑level distribution lines to major transmission corridors — the location of a battery ultimately determines how much benefit it can deliver to the grid and to customers.”
There were 115 megawatts of battery storage operational in New York City at the end of last year, according to Con Edison, and 865 megawatts of projects with interconnection agreements. Peak load in the region is about 10,000 megawatts, meaning that these new projects would meaningfully alter the way the utility serves its customers.
Con Edison has claimed in a regulatory filing that the concentration of projects could lead to “significant impacts from BESS charging on infrastructure upstream of primary feeders,” necessitating the changes to its interconnection process. The city claimed in its filing that the added cost has “understandably chilled ongoing development activity at a time when New York City needs more supply resources capable of serving peak demand.”
When I reached out to the Mayor’s Office of Climate & Environmental Justice about the dispute, I received a statement in return from New York City Chief Climate Officer Louise Yeung: “Expanding battery storage capacity will be critical to New York City’s clean energy future, as extreme climate events continue to strain our grid system,” she said. “The City is working across agencies and communities to ensure battery energy storage projects are deployed safely and can provide reliable power when New Yorkers need it most.”
As for residential solar and storage, it will likely take years for those distributed resources to become a regular part of New York City’s energy landscape. There’s only one fully permitted and approved residential storage system allowed in New York City, which was installed earlier this year by Brooklyn Solar Works. Negotiating approvals with city agencies including the Department of Buildings and the New York City Fire Department took around six years, the company’s vice president of sales, Steve Nelson, told me.
“It’s New York City. We’re expecting there to be some level of bureaucracy and some lift to get that stuff approved,” Nelson said. “But what we also lack is a ready, readily accessible residential battery that meets the criteria that these departments have set.”
All that adds up to both a practical and a political gap for decarbonization, Zarelli told me.
“Batteries are a great way to connect the climate agenda and the affordability agenda, and it’s in the mayor’s control — it’s the regulatory apparatus at FDNY,” he said. “That’s a big near-term play that I think would make a big difference.”
Earlier this year, New York City Councilmember James Gennaro introduced a bill to amend the fire code to relax some battery storage permitting and safety requirements. But that still leaves the city’s decarbonization advocates with many big fish to fry.
“It’s a challenging future that’s still out in front of us, and how to navigate that is really difficult. But it’d be good if we were actually aligned on what our goals were as a society,” Zarelli said.
Rates were up 17% year over year in June, according to the latest Electricity Price Hub update, with another increase on the way.
With higher temperatures come higher electricity bills. Whether through higher seasonal charges or greater usage, Americans across the country were paying more for electricity in June.
In Virginia, the epicenter of the data center boom, the typical household electricity bill was $192 in June, up from $172 in June of last year, according to the latest data from the Heatmap and MIT’s Electricity Price Hub. Rates, meanwhile, were about 18 cents per kilowatt-hour, compared to just over 15 cents in June of last year, a 12% hike. Rates were also up from the end of last year, when they were about 15.5 cents.
The rate increase is largely due to prices set by Virginia’s largest utility, Dominion. Its rates are up 8% so far this year, according to MIT researchers, and 17% over the past 12 months, the result of a base rate increase that took effect at the beginning of the year. The average base rate alone is up 7.5% year over year for the average Dominion customer.
But that’s not all: The fuel portion of the bill is rising $8 a month for the typical customer, Dominion said according to local media reports, as a result of rising costs. The fuel charge went into effect at the beginning of July. Already, Dominion customers are paying about $78 per month for the generation portion of their electricity bill, according to Heatmap-MIT data.
The price hike will likely increase pressure on Dominion as it seeks to sell itself to Florida utility and energy developer NextEra in a $67 billion deal announced in May.
Earlier this week, Virginia's lieutenant governor Ghazala Hashmi sent a detailed letter to the State Corporation Commission, Virginia’s utility regulator, with 64 questions about the proposed merger. She said the deal “carries unprecedented implications for Virginia’s consumers and regulatory landscape.”
Hashmi asked regulators to extend their review of the deal beyond the six-month period mandated by its utility regulations, writing that “forcing this process into the six-month timeline will render an already inadequate period completely unworkable.”
In May, when the deal was announced, NextEra said it would provide over $2 billion of bill credits over two years to Dominion customers in Virginia, North Carolina, and South Carolina, which Dominion executives estimated would add up to $10 per month over the two years.