Portuguese 
English 
Spanish 
7 min of reading 
0 comments 
The expansion of artificial intelligence has increased the electricity consumption of data centers, caused emissions to rise in giants like Google, Microsoft, Amazon, and Meta, and has made it harder to meet climate goals set for 2030, amidst the advancement of natural gas and the pressure for rapid energy
The expansion of artificial intelligence has begun to pressure the climate goals of major technology companies, which are facing increased emissions, accelerated growth in electricity consumption, and greater reliance on natural gas amid the race to build data centers and maintain competitiveness.
Six years ago, Google claimed that by 2030, all of its operations would be powered by electricity generated from clean sources, such as wind and solar energy, in addition to removing from the atmosphere the same volume of pollution that it produced.
Today, the company classifies these goals as an “ambitious project,” while Microsoft maintains the intention to remove more carbon than it emits by 2030, but has begun to describe the effort as “a marathon, not a sprint.”
-
Mercor paid $1.5 million per day for doctors, lawyers, and former Goldman Sachs bankers to teach artificial intelligence to do their jobs, and in 17 months, it went from zero to $500 million in annual revenue while its own contractors accelerated the replacement of their own work.
-
Casio Unveils Moflin, Robotic Pet With Artificial Intelligence Designed To Provide Emotional Comfort And Simulate A Permanent Affectionate Bond
-
Google Maps Revolutionizes GPS Navigation With Gemini Artificial Intelligence, Ask Maps Feature, and Advanced 3D Visualization for Route and Travel Planning
-
China Warns the U.S. of a “Terminator” Style Apocalypse: Military AI Rivalry, Sanctions Against American Startup, and Fear of Algorithms Deciding Who Lives or Dies Place the World on the Brink of a Nightmare That Seemed Like Fiction
The race to implement artificial intelligence has altered the landscape in which these goals were set. Industry giants argue that they need flexibility to build enormous data centers, facilities capable of consuming more energy than entire cities.
Patrick Huang, a senior analyst at Wood Mackenzie, stated that even without an official review of the goals, companies have begun to recognize that they may not be on the expected path. For him, the need to remain competitive forces the industry to use any available type of energy, and increasingly this energy comes from natural gas.
This movement occurs despite technology companies having purchased record volumes of clean energy in 2024 and 2025, according to the Clean Energy Buyers Association. Nevertheless, sustainability reports show that total emissions have increased over the first five years of climate commitments.
Google’s emissions jumped nearly 50% during this period. Amazon’s emissions rose by 33%, Microsoft’s by over 23%, and Meta’s by over 60%.
Artificial intelligence increases pressure on climate goals
The energy burden of data centers helps explain the deterioration of these indicators. In 2024, these structures consumed about 4.6% of all electricity used in the United States, a share that could nearly triple by 2028, according to government estimates.
Some analysts project that electricity consumption in the country could grow by up to 20% in the next decade, with data centers being one of the main factors. At the same time, a backlog of projects awaiting authorization to connect to the power grids and the actions of the Trump administration against renewable sources increase the pressure on companies’ climate goals.
Julie McNamara, associate director of policy for the Climate and Energy Program at the Union of Concerned Scientists, said that each of these factors would already be a significant challenge on its own. In her assessment, the combination of all of them has created a short-term crisis for the system.
Technology companies claim they have been reducing emissions through energy efficiency measures, purchasing renewable energy credits, buying energy from sources that do not emit greenhouse gases, and requiring suppliers to also reduce their emissions. Still, the acceleration of artificial intelligence has increased the demand for electricity at a pace that has begun to strain these efforts.
In 2024, natural gas accounted for more than 40% of the electricity powering data centers in the United States. Globally, coal provided 30% of the energy consumed by these facilities, according to the International Energy Agency.
Race for energy favors the advancement of natural gas
The trend of increasing natural gas shows no signs of retreat. Power companies are planning new plants in various parts of the country to supply data centers, while some technology companies are designing gas plants built exclusively to serve a single facility.
Lori Bird, director of the U.S. Energy Program at the World Resources Institute, stated that companies are seeking to obtain as much energy as possible in the shortest time available. For her, it is an intense race, marked by fierce competition for resources.
Brad Smith, president of Microsoft, told the Associated Press that he is confident in the company’s ability to meet its 2030 goal of removing more carbon dioxide from the atmosphere than it emits. He cited investments in new carbon-free sources, such as nuclear, solar, and hydropower.
In Wisconsin, two new natural gas plants that will supply a Microsoft data center will be offset by investments in solar energy in other parts of the state. In rural Louisiana, three natural gas plants will provide electricity for a massive Meta data center, while the company invests in solar energy in other locations.
Google also claims to invest in wind, hydropower, battery storage, and advanced nuclear energy, although it continues to rely on natural gas. The company plans to purchase electricity from a gas plant that will be built at the Archer Daniels Midland corn processing facility in Decatur, Illinois, with underground capture and storage of carbon dioxide emissions.
To support clean energy goals, these companies rely on power purchase agreements and the acquisition of renewable energy certificates, which support both new and existing sources. However, this model may face new challenges with proposed changes to greenhouse gas reporting rules, which would require regional and temporal matching between the energy source and the operation of the data center.
In practice, this would mean that solar energy credits, for example, could only be applied to daytime operation periods. The change tends to restrict the use of these mechanisms at a time when companies increasingly depend on them to sustain their commitments.
Trump administration measures increase uncertainty in the sector
The energy dispute has become even more complex with President Donald Trump ’s actions against renewable sources. The supply of electricity was already a challenge before his return to office last year, but the situation worsened after the cancellation of subsidies and permits for solar and wind energy projects.
Trump also attacked tax incentives for renewable sources and ordered that several coal-fired power plants scheduled for deactivation remain in operation. Advocates for renewables argue that wind and solar projects can be built more cheaply and quickly than natural gas or nuclear plants.
Rich Powell, executive director of the Clean Energy Buyers Association, stated that many companies set goals counting on the support of federal tax credits for the expansion of wind and solar energy. However, these credits will end in July, after being eliminated by the Republican-controlled Congress and Trump.
Trump, who called climate change a “hoax,” argues that green energy is expensive and unreliable, potentially compromising national energy independence. Powell said that his association made it clear to Congress and the administration that all technologies need to compete on a level playing field, at risk to the accessibility and reliability of energy.
Josh Parker, head of sustainability at Nvidia, stated that artificial intelligence should reduce electricity consumption in the future as it is more efficient than traditional computing. He also stated that restricting energy development could cause the United States to fall behind in the AI race.
Goals for 2030 are viewed with more caution
Jay Dietrich, a sustainability researcher in AI at the Uptime Institute and former leader of emissions target setting at IBM, stated that in 2020, it was very difficult for companies to project current energy needs. Much of the technology and equipment used to train machine learning models, responsible for the majority of electricity consumption in data centers, was still being introduced.
In his assessment, by 2023, technology companies already had a good sense that the scenario would become much more intense and that the numbers would grow rapidly. Dietrich predicts that many companies will extend the timeline of their emissions targets.
This assessment is supported by a Uptime Institute survey from 2025, which recorded a 12% decrease in the number of operators who stated they would meet the carbon neutrality target by 2030, based on the market. Even with the increase in emissions, he believes that the largest companies can still invest in renewable energy and offsets in sufficient volume to meet these goals.
The advancement of artificial intelligence also appears in another measurement. A study by the Rhodium Group pointed to the technology as partially responsible for the 2.4% increase in fossil fuel emissions in the United States last year.
Although some new gas plants replace more polluting coal plants, the return on investment takes about 30 years. This prolongs the transition to clean energy at a time when the United Nations Environment Program is already warning that countries with high emissions are unlikely to meet their own reduction targets.
McNamara stated that the surge in electricity demand from data centers has turned a challenge into a declared crisis.
For her, artificial intelligence is associated with a significant increase in dependence on fossil fuels under the oversight of technology companies and as a direct result of their actions.
Seja o primeiro a reagir!