×
AI’s energy impact and the urgent debate we’re ignoring
Written by
Published on
Join our daily newsletter for breaking news, product launches and deals, research breakdowns, and other industry-leading AI coverage
Join Now

AI’s growing energy demand and its implications: The rapid expansion of artificial intelligence (AI) technology is creating a significant surge in energy consumption, raising concerns about its impact on electricity costs and the environment.

  • Big Tech companies like Amazon and Google are receiving preferential electricity deals from utility companies for their data centers, which are essential for AI operations but consume vast amounts of energy.
  • These sweetheart deals often result in higher costs for everyday consumers, as utility companies seek to recoup their investments and maintain profits.
  • The energy-intensive nature of AI and data centers is putting strain on the existing electricity infrastructure, potentially leading to the need for new power plants and transmission lines.

The monopoly utility system and its consequences: America’s outdated monopoly utility system is exacerbating the problems associated with AI’s growing energy demand, leading to unfair pricing and environmental concerns.

  • Utility companies, such as Dominion Energy, are incentivized to invest in new power plants and transmission lines due to guaranteed profits, often at the expense of consumers.
  • Regulators have allowed utilities to impose higher fixed charges on customers, particularly those with solar panels, creating a disparity between efficient users and energy-intensive consumers.
  • The pricing structure often favors industrial customers over residential and small business users, despite the higher costs associated with serving residential customers.

Economic incentives and subsidies: Many states are providing significant economic incentives for data center development, despite limited local economic benefits.

  • A 2016 report found that 27 states offered economic incentives for data centers, even though these facilities often purchase little hardware locally and create few jobs.
  • Eight states have gone further by exempting data centers’ electricity purchases from sales taxes.
  • In one extreme case, an Amazon Web Services facility in Ohio received $93 million in subsidies for creating just 120 jobs, equating to over $750,000 per job.

Historical context and ongoing issues: The current problems with monopoly utilities are part of a long-standing pattern of issues in the energy sector.

  • The 1920s saw costly speculation that required massive federal reforms.
  • In the 1970s, poor planning led to a significant increase in electricity rates, with average rates more than quadrupling between 1970 and 1983.
  • Utilities have consistently under-invested in energy efficiency due to a lack of profitability for shareholders, despite it being the most cost-effective way to meet electricity demand.

Recent monopoly-induced disasters: Several recent incidents highlight the ongoing problems with the monopoly utility system.

  • Ohio’s FirstEnergy spent $60 million bribing public officials to win billions in fraudulent subsidies for aging power plants, with penalties recovering only a quarter of the illegally obtained funds.
  • California’s Pacific Gas & Electric’s prioritization of profits over maintenance led to multiple devastating wildfires, resulting in dozens of deaths and billions in property damage.

Environmental concerns and fossil fuel dependence: The expansion of data centers is being used to justify investments in fossil fuel infrastructure, raising environmental concerns.

  • In Georgia, utility lobbyists are using reports of high load growth from data centers to justify significant investments in new natural gas generation capacity.
  • These investments in gas infrastructure put customers at financial risk, as demonstrated by the price spikes during Winter Storm Uri in Oklahoma.
  • The continued reliance on fossil fuels for data centers contradicts many tech companies’ public commitments to sustainability and clean energy.

Proposed solutions and policy recommendations: To address the challenges posed by AI’s energy demand and monopoly utilities, several policy measures are suggested.

  • The Federal Trade Commission should consider breaking up Big Tech firms to reduce their power in negotiating sweetheart deals and investigate anti-competitive behavior in the utility sector.
  • State legislatures should implement regulations requiring data centers to use state-of-the-art technology for power reduction and be supplied entirely by renewable electricity.
  • States should introduce competition in power generation and reduce barriers to renewable energy grid connections to break the utility monopoly grip.
  • Congress should pass legislation to remove the antitrust shield for private utility companies.
  • The federal government should collect data on interconnection timelines and costs to uncover utility efforts to hinder clean energy construction.

Broader implications: The intersection of AI’s energy demand and the monopoly utility system presents both challenges and opportunities for reform.

  • Addressing these issues could lead to a more equitable, efficient, and environmentally friendly energy system that better serves consumers and the climate.
  • By breaking the monopoly grip over both the digital economy and the electricity system, there is potential to create a more balanced approach to technological advancement and energy consumption.
  • The focus on data centers and AI energy use could serve as a catalyst for broader reforms in the energy sector, potentially leading to a more sustainable and consumer-friendly electricity system.
We aren’t ready for AI: The energy debate we need to have

Recent News

AI-powered computers are adding more time to workers’ tasks, but there’s a catch

Early AI PC adopters report spending more time on tasks than traditional computer users, signaling growing pains in the technology's implementation.

The global bootcamp that teaches intensive AI safety programming classes

Global bootcamp program trains next wave of AI safety professionals through intensive 10-day courses funded by Open Philanthropy.

‘Anti-scale’ and how to save journalism in an automated world

Struggling news organizations seek to balance AI adoption with growing public distrust, as the industry pivots toward community-focused journalism over content volume.