How AI Enables Fossil Fuel Expansion and Lock-In
AI Accelerates the Oil and Gas Industry’s Production, Exacerbates the Climate Crisis, and Strengthens Anti-Democratic Forces While Hiding Behind the Myth of Efficiency
by Stella Levantesi
She worked for Microsoft, founded the program that invested in sustainability projects in the communities that host data centers, and co-founded the company’s employee engagement program for sustainability that included over 10,000 members. Then, one day in early 2024, after nearly a decade at Microsoft and a troubling discovery, Holly Alpine resigned.
“We discovered contracts between Microsoft and fossil fuel companies specifically and explicitly helping them discover more oil, produce, extract, refine, transport more. Every step of the value chain was being optimized by Microsoft technology,”
explained Holly Alpine, co-founder of the Enabled Emissions campaign, in an interview with Voice Over.
“What we saw at Microsoft – and what you see across the industry's largest providers – is that AI is being sold into fossil fuel operations at scale. But the public conversation, the environmental movement, the regulatory energy – almost all of it is focused on the electricity the data centers consume,” Alpine explained.
This framing is missing the larger story, Alpine argues. The link between AI and the climate crisis goes far beyond the energy consumption of data centers.
“When AI makes fossil fuel extraction cheaper and more viable, it expands production that wouldn’t otherwise have happened. Those are the emissions that dwarf everything else, and they're almost entirely absent from how we're governing AI’s climate impact right now,” she added.
The Fossil Fuel Accelerator
Microsoft AI, Google Cloud, NVIDIA, AIQ, Baker Hughes C3 AI, AiPSO,Mistral… to each AI technology its own partnership agreement with the oil and gas industry.
More than 92 percent of fossil fuel companies are investing in AI or plan to do so, according to a survey by consulting firm EY.
The oil and gas sector’s spending on AI will increase from about $4 billion in 2025 to $13.4 billion in 2029, a 235 percent rise, according to an analysis by Deloitte.
Microsoft, which holds more than half of all cloud service partnerships with the fossil fuel industry, and Amazon are currently “leaders in the oil and gas space,” according to a report by Barclays titled “Frac to the Future; Oil’s Digital Rebirth.”
Amazon Web Services (AWS), “the world’s most comprehensive and broadly adopted cloud” has agreements with many of the major oil and gas companies, including BP, Exxon Mobil, and Shell, and earns billions by selling customized AI services to companies in the fossil fuel sector, helping them extract more oil and gas.
According to Shell’s statements, AWS’s AI models make it possible to identify and develop new oil and gas fields much more quickly, and facilitate exploration in some of the most remote regions of the planet, a report by Greenpeace Germany revealed.
“This is exactly what we anticipated, that using the [AI] infrastructure makes this dirty business model more profitable and even more harmful, and pushes forward environmental destruction,” Linda Klapdor, AI and Big Tech expert at Greenpeace Germany, emphasized in an interview with Voice Over.
Today, AI is used at every stage of the fossil fuel supply chain, from exploration to production, from transportation to distribution.
“Advanced technology is very good at taking large amounts of data and turning it into insights and action. The oil and gas industry has a century of data, very large volumes from multiple sources.
You’ve got your wells, you’ve got your seismic data, you’ve got your production information, and that can all be fed into simulation models that can turn all the data into insights into where to drill, how to drill, and how to do it cheaper and faster,”
Alpine explained.
AI increases the feasibility of oil and gas operations that, previously, were impossible.
For example, AI has helped the U.S. oil producer Devon drill in areas where it was previously unfeasible, said Chief Technical Officer Trey Lowe in an interview with Reuters.
The AI solutions provider SparkCognition claims that its patented technology will make it possible to identify previously unexplored fossil fuel resources on the ocean floor in just nine days instead of nine months, according to a report by the organization Global Witness.
ExxonMobil has partnered with IBM on an AI solution that is said to have accelerated its ability to design drilling plans for new wells by two months.
“All this really helps to speed up development timelines and help where there are projects that would otherwise be too expensive or too difficult to move forward […] These projects are key to the companies’ profits and also key for the fossil fuel industry’s continued success” Alpine stated.
Oil and Gas’ Future Is “Brighter Than Ever”
Messaging from the oil and gas industry and Big Tech is rife with claims about the benefits of AI for fossil fuel operations.
With Microsoft Azure, “the future of oil and gas exploration and production is brighter than ever,” says a LinkedIn post by a former product manager at Microsoft who helped the company develop AI that was later sold to oil companies.
“That was the nail in the coffin to leave the company [Microsoft]. That really sums it up pretty well [...] I loved my work there, I loved my colleagues [...] but I became very concerned,” Alpine said.
“Companies want to be seen as having their lifeline extend into the future. I think the basic intuition that most people have is that digging stuff out of the ground and burning it, is not the future,”
Duncan Meisel, a climate communications expert and executive director of Clean Creatives — a global campaign for PR professionals and firms who aim to pledge to stop working with fossil fuel clients — said in an interview with Voice Over.
Communication surrounding AI and fossil fuels is also loaded with promises of profit for industries.
Microsoft talks about increasing “Permian profitability” for ExxonMobil through their partnership: Permian application, launched in 2019, will generate billions of dollars over the next decade by promoting “capital efficiency.”
A former Microsoft engineer, who traveled to Kazakhstan to integrate the Tengiz oil field with the company’s AI technology,reported an increase in daily oil production from 600,000 barrels to 1 million thanks to the multimillion-dollar partnership with Chevron.
Ann Davies, senior vice president of wells at the oil and gas multinational BP, stated that, thanks to AI, they are “able to drill more wells per year and have a better capital allocation,” Reuters reported.
What was previously beyond the reach of the oil and gas industry is now achievable thanks to AI—and Big Tech and Big Oil companies continue to profit.
AI and digitalization are advancing at an unstoppable pace and, between 2026 and 2030, they could reach a cumulative value of $500 billion for oil and gas companies, according to a May 2026 analysis by the consulting firm Rystad Energy.
“It’s not just about climate, it’s also about human rights and political integrity and democratic values. And it seems there’s not any responsibility taken. And I think that’s very concerning,” Klapdor emphasized.
Before leaving Microsoft, Alpine spent many years trying to bring about change from the inside.
“We worked for many years internally, wrote up a memo that had specific concerns and recommendations, and met with the president and other senior leaders. We got a lot of promises but really those promises never came to fruition. We just never got the transparency, the accountability, the responses and, really, the action that we were requesting,” Alpine explained.
After leaving the company, Alpine co-founded the Enabled Emissions campaign, which aims to hold Big Tech accountable for AI enabled fossil fuel expansion, along with her husband Will, a technology and climate expert and former Microsoft employee like Holly.
“We believe our research and the research that other experts have done shows that we would be going in a different trajectory worldwide in terms of emissions, if not for this technology keeping the fossil fuel industry afloat. I would call that [climate] obstruction and delay, definitionally,” Alpine emphasized.
Behind the Mask of “Efficiency”
Digital technology, including AI, says Shell on its website, increases the “efficiency” of exploration and production.
Exxon Mobil talks about “pairing” its “domain knowledge” with AI to develop more “efficient” processes.
“We increase the efficiency and precision of our operations through the use of cutting-edge digital technologies, such as supercomputing and artificial intelligence,” reads Eni’s website.
BP’s web page about AI and the energy system contains the words “efficiency” and “efficient” a total of 14 times.
“This conversation about efficiency is subtle greenwashing. [Companies] use the idea of efficiency because it feels ‘clean’. The assumption is that oil and gas production is inevitable, that we’re always going to use oil and gas and, therefore, if it’s more efficient, we’re better off,”
Meisel explained.
“What they’re just doing when they’re increasing their ‘efficiency’ is they’re lowering their costs so they can produce something at higher rates for the same amount of capital investment. So really, they’re just trying to boost their profit margins and sell it to you as some sort of environmental insight […] The problem we face is using oil and gas, not how efficient it is to get the oil and gas out of the ground,” Meisel underscored.
The language says ‘efficiency,’ ‘optimization,’ ‘innovation,’ but the meaning is very simple: more gas and more oil, and at ever-increasing rates.
“We’ve continued to have more assets that are running at a longer time, so this is all around just producing more oil or gas,” Russell Robinson, a deputy program manager of facilities and operations at Chevron, which uses AI-integrated drones to monitor extraction facilities in Texas and Colorado, said in an interview with Reuters.
“By saying it makes them more efficient, what they [oil and gas companies] are actually saying is, we can exploit the planetary resources even faster and cheaper,”
Klapdor pointed out.
Big Tech’s “vague-washing”
Experts refer to AI as a lifeline for the fossil fuel industry, but even more so than a lifeline, it appears to be a vicious cycle: artificial intelligence drives more oil and gas, which, in turn, powers the immense, resource-draining infrastructure of data centers.
“Data centers are hyper local infrastructures so they need huge amounts of energy in a very specific place, which usually the grid cannot handle. So what they do is they put onsite power generation which, usually, is not wind turbines or solar panels, but gas turbines,” Klapdor explained.
“In Germany, there are cases where they installed gas power plants because the local grid can’t handle [the data centers]. It’s very alarming,” she added.
Data centers are responsible for about 1.5 percent of total annual electricity consumption worldwide. This figure has grown by 12 percent annually over the past five years, and projections indicate that it is set to more than double by 2030 —from 415 terawatt-hours (TWh) to 945 TWh — primarily for artificial intelligence-related purposes.
In Texas, where Meisel lives, a large data center is under construction.
“The reason there’s a data center build out here is that we have a lot of gas and it’s very easily accessible…Fossil fuel for electricity consumption would be going down dramatically if it wasn’t for AI in the US and the global north,” Meisel said.
In some cases, fossil fuel companies are pitching gas, oil, and coal infrastructure to power AI with carbon capture and storage (CCS) facilities. CCS is promoted to fuel the perception that the oil and gas industry and AI are “green” but, in reality, it allows fossil fuel lock-in, allows for continued pollution, and delays the transition to renewable energy, according to an analysis by the Center for International Environmental Law (CIEL).
Major Big Tech companies tout the climate benefits of AI, but according to a new report, these claims are a “hoax.” Out of 154 statements claiming that AI will benefit the climate—including claims from Google, Microsoft, and the International Energy Agency—only 26 percent cite published academic papers, and 36 percent cite no evidence at all, the study found.
“The promises of planet-saving tech remain hollow, while AI data centres breathe life into coal and gas every day,”
said climate and energy analyst Ketan Joshi, who led the research and authored the report published by a coalition of organizations, including Beyond Fossil Fuels and Climate Action Against Disinformation (CAAD).
The report also mentions “tactical vagueness” as a new form of greenwashing and cites journalist Karen Hao’s book “Empire of AI,” which recounts how the phrase “artificial intelligence” was originally coined as a marketing term in the 1950s and later used to describe a broad set of technologies.
“People think that to get any benefit from AI in general, which is a huge category of technologies, they somehow have to accept the most environmentally egregious form of AI. I really reject that premise. The term AI is so vague that it’s like the term transportation. You could be talking about a bicycle or a rocket…But that doesn’t mean that the benefit from using the bicycle also justifies us building the rocket,” Hao pointed out in an interview.
Across these sectors, one of the main challenges remains the lack of transparency.
“There is no accountability and transparency…That’s what we’re mainly trying to do [with Enabled Emissions] is start with this baseline of transparency and to have the public and shareholders and policy makers understand what’s going on and the impacts of it,” Alpine highlighted.
“What are we collectively going to decide to do? I think more than ever, the future of the energy system and our technology is a question that we have to decide. I think that’s why you see so much investment in ads and PR and political engagement from the oil and gas industry right now, and why you see all these big tech companies showing up at Trump events,” Meisel concluded.
“They know that’s where all the power and the future lies, in that collective political choice.”
