On Tuesday, the US Senate voted overwhelmingly to remove a controversial provision in the One Big Beautiful Bill Act that sought to impose a ten-year moratorium on state-level regulation of artificial intelligence.
The vote, which passed with a resounding 99-1 margin, marks a significant shift in the ongoing discourse surrounding AI policy in the United States.
The provision, which was backed by President Donald Trump’s administration, aimed to prevent states from creating inconsistent regulations while the federal government formulates its own framework for AI.
However, the Senate’s decision reflects growing concerns about the potential consequences of delaying state-level regulation in a sector that is rapidly evolving.
One major challenge highlighted by this vote is the divergence in priorities between the federal executive branch and state governments, particularly regarding AI infrastructure development.
While the benefits of AI are broad and often speculative, immediate local impacts such as public health risks and increased energy costs associated with the expansion of data centers are tangible and immediate.
In fact, local opposition has already thwarted approximately $64 billion worth of data center projects within the last two years, complicating the administration’s efforts to maintain U.S. leadership in the global AI landscape.
The push for rapid data center expansion, as advocated by the Trump administration, requires greater collaboration between federal and state governments to avoid inefficiencies and excessive financial losses while ensuring AI serves the public good.
Virginia is currently leading the charge in data center capacity, boasting 5.9 gigawatts, with an additional 1.8 gigawatts slated to come online soon.
The state plans a staggering 15.4 gigawatts of new capacity, much of which is tied to AI advancements.
This data center boom has created substantial economic benefits for Virginia, contributing around 74,000 jobs, generating $5.5 billion in labor income, and adding $9.1 billion to the state’s gross domestic product annually.
However, the rapid expansion is built upon speculative demand, raising concerns regarding a potential data center bubble.
A recent study indicates that the infrastructure required to support new data centers could result in higher utility costs, with an average household projected to see an increase of between fourteen to thirty-seven dollars monthly by 2040.
Public health implications must also be considered, as increases in pollution and emissions from data centers could disproportionately affect low-income communities, as noted in a 2024 McKinsey study.
Similar to historical analogies such as the California Gold Rush, which resulted in widespread displacement and environmental degradation, the current data center boom may lead to adverse social and economic consequences.
Looking internationally, China’s Eastern Data, Western Compute (EDWC) initiative serves as both an instructive example and a warning.
Launched in 2022, EDWC aimed to establish ten national data center clusters in economically deprived but resource-rich western regions.
While ambitious, the initiative faces significant financial challenges, costing an estimated 400 billion renminbi ($55 billion) annually, despite turnover figures lagging behind at 150 billion renminbi ($20 billion).
The initiative has witnessed rapid declines in data center service prices as supply outpaces demand, resulting in high vacancy rates in newly established centers.
Moreover, AI-driven data centers incur higher operational costs—potentially double that of traditional facilities—due to heavy investments in specialized hardware and the necessity for constant operational readiness.
As articulated by a Chinese tech outlet, the situation begs the question: “If a highway is built but only a few hundred cars are on the road each year, who will bear the amortized costs?”
In light of these complications, the United States federal government must step in to mitigate risks associated with oversupply.
This includes potentially establishing an AI Data Center Monitoring Initiative (AI-DCMI) that would work alongside federal entities such as the Department of Energy, the Federal Energy Regulatory Commission, and state energy offices.
The AI-DCMI would collate data and develop metrics that assess the impact of data center operations on local communities, public health, and market competitiveness, while providing clarity on the expected utilization of AI infrastructure.
Investment incentives related to data center projects should be connected to reporting requirements under this initiative.
While the AI-DCMI would not have regulatory authority, it could serve as a vital resource for both state and federal governments to better understand the current risks of AI data center investments.
The recent Senate vote underscores the vital need for coordinated efforts to ensure the development of AI infrastructure is not only financially viable but also aligned with the broader interests of the American public.
As the next phase of AI infrastructure development unfolds, the stakes are high, and without enhanced collaboration, the U.S. risks detrimental financial implications while rivals advance.
Moving forward, careful planning and state-federal coordination will be essential in harnessing AI’s potential to benefit society as a whole.
image source from：atlanticcouncil
