Artificial intelligence (AI) is reshaping industries at a breakneck pace. AI isn’t just about computing; it’s about electric grids, data centers, and the policies that dictate where resources flow. The people who control this infrastructure have as much influence over AI’s future as the engineers building it.
While investors salivate over potential profits in the blossoming industry, AI’s insatiable demand for energy is pushing U.S. infrastructure to the brink. To accommodate its needs, President Donald Trump is hyperfocused on expanding that infrastructure.
The Trump family is also privately betting big on AI’s energy-guzzling backbone: data centers. The data center market is expected to be worth nearly $500 billion globally by 2029.
Eric and Donald Trump Jr. are investing in the industry via Dominari Holdings. The two, who were also named to Dominari’s advisory board, touted their “strategic insight” in the AI and data center sector, Forbes reports.
Their investments have sparked concerns about potential conflicts of interest, as their father is positioned to direct government funds to the industry, alter policies that benefit it, and use the office to attract investments in it that could potentially benefit his family and those associated with them.
In January, Trump announced a $20 billion investment in data centers by an Emirati company helmed by a man the Associated Press described as a “close business partner” of the family.
Trump has bragged that he intends to get investments of $1 billion or more in the industry.
Some observers wonder whether these investments are necessary to meet the country’s energy needs, or more about securing a financial windfall as AI companies drive electricity demand through the roof.
The intersection of AI, energy policy, and political influence could permanently reshape the nation’s power grid and alter the flow of power resources at large.
Accommodating AI’s energy demands
Every ChatGPT query, Midjourney render, and autonomous vehicle decision runs on an expanding network of data centers. Training an AI model like GPT-4 requires the same amount of electricity a small city might use in a month. A recent report from the International Energy Agency (IEA) predicts that power demand from AI, cryptocurrency, and data centers could double by 2026. Data centers’ projected 2026 electricity consumption alone is equivalent to that of Japan. The IEA projects that U.S. data centers’ electricity consumption will rise from about 460 terawatt-hours (TWh) in 2022 to more than 1,000 TWh in 2026, accounting for 6% of total electricity demand.
That growth is fueling a national energy infrastructure crisis.
The strain is already visible. In Texas, where data center construction is booming, the state’s largest power grid operator warned that energy demand could exceed supply within the decade.
The problem isn’t limited to the U.S. In Ireland, power shortages linked to the rapid expansion of data centers forced government regulators to consider energy rationing. If AI continues on its current trajectory, regions with high data center density could face similar energy prioritization decisions. In this scenario, private homes and businesses would compete against tech giants for electricity.
“After 15 or more years of stagnant electricity demand, AI is now increasing the need for more electricity across the country,” Constantine “Costa” Samaras told the Daily Dot. Samaras is the director of the Carnegie Mellon University Scott Institute for Energy Innovation and former principal assistant director for energy and was chief advisor for the Clean Energy Transition in the White House Office of Science and Technology Policy under the Biden administration.
“While data centers currently consume about 5% of U.S. electricity, the biggest risk is failing to address AI’s energy demands. If we can’t get it right for AI, we won’t get it right for broader electrification that can help address climate change,” Samaras added.
The Trumps, Dominari, and the data center gold rush
The Trump family has never been one to sit out a lucrative opportunity. AI is a potential gold mine.
Dominari Holdings recently launched American Data Centers to develop high-performance computing data centers for the AI infrastructure market. Dominari holds a 32% stake in this endeavor.
Dominari’s investments in this space could provide the Trump family with not only financial returns but also influence over the future of AI innovation.
As the global AI race intensifies, the question of who controls the infrastructure that powers the technology becomes more critical. AI’s enormous growth is skyrocketing energy use, stretching the nation’s ability to keep up.
“Companies and researchers should always push to make data centers more efficient,” Costa Samaras said. “But when data centers become more efficient, companies just add more computing, and energy use climbs again.
Last year, the Washington Post reported that the U.S. is running out of power, in part due to AI’s demands. Days after taking office, Trump declared a national emergency over “inadequate” energy supply.
Energy providers, such as fossil fuel companies, could stand to gain from Trump’s focus on increasing energy output. The Trump family could also benefit from his moves that keep power supply up and costs down for the energy-hungry AI industry.
This confluence of political power, technological infrastructure, and financial interest highlights the increasing intersection of business and politics in sectors shaping the future of society.
Dominari Holdings and the White House did not respond for comment.
The ultimate insider
David Sacks is a renowned tech investor who serves as the White House AI and Crypto Czar, and also leads the President’s Council of Advisors on Science and Technology. He cofounded PayPal alongside AI advocates Elon Musk and Peter Thiel. Sacks invests in companies owned by both, including Thiel’s AI outfit Palantir Technologies. Last year, he launched an AI-powered generative work chat, Glue, to compete with Slack and Microsoft Teams. A spokesperson for his venture capital firm also confirmed that he remains in his position there.
Sacks’ appointment to a prominent position in the Trump administration raised concerns that he is uniquely positioned to shape policies that benefit him and his business allies. He’s invested significantly in AI, though the extent of his investments are unknown. According to Bloomberg, Sacks’s government role is part-time, allowing him to avoid divesting his investments or publicly disclosing his financial holdings.
This dual capacity is a microcosm of the larger issue at hand: the potential for private interests with skin in the game to dictate public policy in AI that enriches themselves. Some wonder whether, when Sacks or Trump advocate a particular policy or public investment regarding AI, they’re doing so for the good of the country or for their own benefit.
Sacks has denied being compromised by conflicts of interest. Last month, the White House issued a memorandum stating that Sacks had disclosed his financial holdings and, after review, it “determined that the financial interests … are not so substantial as to be deemed likely to affect the integrity of [Sacks’] services to the Government.”
In a recent discussion, Sacks highlighted that access to energy is the number one constraint on AI’s growth.
Sacks believes that whoever controls AI’s energy supply effectively controls AI itself. To that end, he has pushed for aggressive expansion in power generation and favors policies that prioritize infrastructure investment.
Skeptics wonder whether these efforts are less about securing national AI dominance and more about ensuring that private investors—like Sacks and the Trumps—profit from the data center boom. To some, the need to rapidly expand infrastructure to accommodate it raises questions about the long-term sustainability and equity of AI’s growth.
While addressing immediate energy needs are crucial, a singular focus on increasing power generation to benefit one industry can risk overlooking the need for balanced and responsible development.
Critics argue that prioritizing speed and scale may lead to environmental degradation, exacerbate existing inequalities, and ultimately undermine the long-term viability of AI technologies. They say that for a more nuanced approach that considers environmental impact, social equity, and long-term sustainability.
A geopolitical AI race
The geopolitical AI race between the U.S. and China is accelerating efforts to increase energy capacity.
China has poured resources into AI, semiconductor manufacturing, and subsidized energy infrastructure.
China’s state-backed approach gives it a competitive edge in securing energy for AI. In regions like Inner Mongolia and Sichuan, where electricity is abundant due to coal reserves and hydropower, the government has fast-tracked AI infrastructure projects. These regions have become AI powerhouses, hosting massive data centers that fuel machine learning and cloud services.
The U.S. and China are taking vastly different approaches to managing AI’s energy needs.
In 2023, China allocated over $10 billion in incentives to support AI-related energy consumption.
The U.S., by contrast, is dealing with power grid constraints that threaten to slow AI expansion, forcing policymakers to balance growth with energy availability.
This disparity raises concerns about long-term competitiveness. If China keeps energy costs low for AI while the U.S. grapples with energy shortages that stymy the industry, Chinese AI companies could gain a significant edge in the global tech race.
That’s created further incentive for private investors in the U.S., including Trump-backed Dominari, to move quickly. Building energy-intensive data centers could be the key to long-term AI dominance.
More than money
The push to beat China isn’t just about economics.
The U.S. is in a technological arms race with its rival. Beijing has been rapidly deploying AI-driven defense and surveillance systems, fueling fears that the U.S. must match China’s pace to maintain its strategic advantage. It was reported as early as 2017 by the New York Times that Beijing was aiming to be the leader in AI by 2030. China’s “Made in China 2025” plan directs significant state funding to the AI industry.
As detailed in a Center for Strategic & International Studies analysis of China’s pursuit of defense technologies, the nation’s focus on AI-driven military applications, coupled with its substantial investments in related technologies, emphasizes its commitment to achieving technological superiority.
The Department of Defense has expressed growing concern over China’s advancements in AI-powered military applications, including autonomous weapons systems and advanced surveillance technologies.
A Pentagon report published in December indicates that China is rapidly closing the technological gap with the U.S. in critical AI domains, potentially challenging U.S. military superiority. It stated that by 2030, China aims to deploy various capabilities related to “algorithmic warfare” and “network-centric warfare,” with varying degrees of human-machine interaction. Chinese Communist Party leadership sees AI and machine learning as tools to bolster information gathering, surveillance, and reconnaissance, while also facilitating the development of novel defense systems, notably autonomous and precision-strike weaponry.
This competitive landscape adds urgency to U.S. efforts to bolster its energy infrastructure, while also raising questions about the balance between national security and private sector interests.
As U.S. policymakers scramble to keep up, private investors like the Trumps are positioning themselves at the nexus of AI growth and energy scarcity, profiting from both political leverage and market forces.
Shaping the future of U.S. energy policy
As AI reshapes industries and data centers proliferate, U.S. energy policy is undergoing a fundamental shift. The government is now weighing new energy incentives aimed at fast-tracking grid expansions and AI-specific infrastructure projects. Some lawmakers are pushing for regulatory rollbacks that would allow faster construction of nuclear plants and natural gas facilities to support AI.
“The U.S. will never regret investing in a cleaner, more efficient, and more reliable grid,” Samaras said. “A smarter and more resilient grid benefits consumers, companies, and AI innovation alike.”
Meanwhile, environmental advocates warn that unchecked AI-driven energy consumption could set back climate goals. Many of the largest AI firms, including OpenAI and Google DeepMind, claim they are committed to sustainable AI, but the reality is that most of the current power comes from fossil fuels.
“I’m not an expert in that area. I’m not going to pretend to be,” David Sacks said of climate change in a 2022 episode of his podcast. “I do think we can’t save the planet by destroying the economy. And it seems to me that too many of these ‘save the planet’ people want to take reckless, extreme actions that would wreck our economy.”
The AI development race has fueled what many view as unsustainable growth in energy consumption. Some industry leaders suggest that improved efficiency could offset AI’s power needs. Samaras is skeptical that this alone could solve the problem.
“Companies and researchers should always push to make data centers more efficient,” Samaras explained. “But when data centers become more efficient, companies just add more computing, and energy use climbs again. Efficiency is necessary, but it’s not sufficient to solve the challenge.”
As AI’s energy consumption skyrockets, it remains uncertain whether it will push the U.S. toward renewable energy or deepen reliance on fossil fuels. According to projections from McKinsey & Company, U.S. data center electricity consumption, driven significantly by AI, is expected to rise from 147 terawatt-hours (TWh) in 2023 to a staggering 606 TWh by 2030, representing 11.7% of total U.S. power demand. This dramatic increase highlights the need for strategic energy planning.
Who controls the future?
AI isn’t just a technological revolution—it’s a political and economic one. The sheer scale of its energy consumption is forcing the U.S. to make hard choices about infrastructure, power generation, and national security.
These choices are increasingly obscured by the potential for conflicts of interest. When major political figures, like Sacks and the Trumps, are financially entangled with the burgeoning industry, the specter of self-serving legislation and regulatory capture looms large, threatening to distort priorities and undermine public trust.
Geopolitically, AI’s energy demands create new resource wars and strategic dependencies. Economically, it risks deepening inequality, with a few controlling the energy and data needed to fuel AI’s future.
“It is up to us and the policies we choose over the next five years,” Samaras said. “We can innovate for a cleaner, more affordable grid while incorporating AI’s energy demands, or we can do nothing—and face higher electricity prices, more pollution, and strain on the grid.”
No matter which path the country chooses, those who’ve invested in AI stand to profit handsomely—the question is at what cost to the rest of us.
Send Hi-Res story tips and suggestions here.
The internet is chaotic—but we’ll break it down for you in one daily email. Sign up for the Daily Dot’s web_crawlr newsletter here to get the best (and worst) of the internet straight into your inbox.
