In recent years, the rapid iteration of artificial intelligence (AI) technology has driven the rapid expansion of data center scale, and the electricity consumption of data centers in the United States has shown an exponential growth trend. The public's dissatisfaction with the high energy consumption of data centers driving up electricity bills is becoming increasingly prominent. In this context, US President Trump recently officially announced a new policy targeting global leading data centers and AI companies, attempting to resolve the contradiction between this livelihood pain point and industrial development.

According to this policy, tech giants such as Amazon, Google, Meta, and Microsoft need to build, introduce, or purchase their own power supply for the construction of new AI data centers, in order to avoid the surge in electricity demand in the AI industry that will drive up the electricity bills of the American people. This policy is seen as a key measure by the Trump administration to balance the development of the AI industry with the cost of electricity for people's livelihoods.

Balancing industrial development and people's livelihood burden

Trump said that the existing power grid in the United States is severely aging and unable to cope with the surge in electricity demand from AI data centers. "We are telling major technology companies that they have an obligation to meet their own electricity needs. They can build their own power plants in factories so that no one's electricity prices are pushed up."

White House spokesperson Taylor Rogers further explained that the core goal of the plan is to protect consumers from the impact of rising electricity costs, while promoting the development of the AI industry and safeguarding the electricity rights of ordinary people. This statement also confirms the dual orientation of the policy - neither hindering the global leading position of the US AI industry, nor responding to public concerns about rising electricity bills.

In fact, the aging problem of the US power grid has long been highlighted, with severe shortages of transmission lines in some areas. At the same time, the construction of data centers in the United States has shown explosive growth, further exacerbating the pressure on the power grid. Some of the super large data centers consume enough electricity per year to supply 200,000 households, and their centralized layout not only brings huge loads to the regional power grid, but also directly leads to the continuous rise of electricity bills in some areas. As the region with the highest density of data centers in the world, the 329 data centers in Loudon County, Virginia consume a quarter of the state's electricity, and local electricity bills are expected to rise by over 50% by 2039, which has become a direct cause of public dissatisfaction.

The North American Electric Reliability Council (NERC) predicts that with the rapid development of industries such as AI and cryptocurrency, US electricity demand may reach its highest level since 2016 in the summer of 2026. If grid infrastructure is not upgraded in a timely manner, it may trigger large-scale power outages.

The landing effect is uncertain

Despite the clear intention of the Trump administration's new policy, it still faces many practical obstacles in the actual implementation process.

Wu Zewei, a special researcher at Su Shang Bank, provided a detailed interpretation of this. He pointed out that power projects usually involve complex approval processes, site selection restrictions, and environmental assessments, with construction periods often lasting several years, which is mismatched with the investment pace of the artificial intelligence industry. In terms of funding, the construction of power plants requires high upfront capital investment and bears long-term recovery risks, which puts higher demands on the company's cash flow and financing capabilities. In addition, fuel supply, equipment procurement, and grid connection technology standards also constitute uncertain factors. If there is a lack of mature supporting facilities and professional operational capabilities, enterprises entering the power generation field across borders may face problems such as rising management costs and lower than expected efficiency.

From the perspective of enterprise operation, self-built electricity will significantly increase the cost and capital pressure of enterprises. Wu Zewei further added that in addition to the investment in construction itself, enterprises also need to bear continuous costs such as maintenance, fuel, and safety guarantees, making their capital structure more heavy. Against the backdrop of slowing growth in computing power demand or intense market price competition, cost pressures may compress profit margins and slow down some expansion plans. If a company chooses to pass on the new costs to customers, the rental price of computing power may increase, which in turn may affect the pace of downstream innovation. However, Wu Zewei also pointed out that in the long run, if self-built power sources improve power supply certainty and stability, they may also provide a more controllable cost structure for large-scale projects.

It is worth noting that even if this policy is ultimately implemented, the stability of US electricity prices still faces challenges. Wu Zewei believes that data center loads are highly concentrated and operate continuously. Even if they are self used, they often need to be connected to the power grid as backup and peak shaving support, which will still put pressure on the system during peak hours. If a large number of enterprises simultaneously lay out their own power plants, it may change the load structure of the power grid, drive up dispatch and maintenance costs, and then transmit them to the whole society through the electricity pricing mechanism. The aging of the US power grid and prominent bottlenecks in the transmission and distribution system may exacerbate regional imbalances if self built power sources lack unified planning.

Overall, Trump's new policy attempts to solve the contradiction between the pressure on the power grid and the burden on people's livelihoods through the "corporate responsibility" approach, but it has not fundamentally addressed core issues such as the aging of the US power grid and the unreasonable layout of power supply. In the future, how to balance enterprise costs, power grid safety, livelihood burden, and environmental protection needs remains a problem that needs to be solved in the implementation process of this policy.