AI Productivity Gains in Fossil Fuels Outweigh Clean Energy Benefits, Study Finds

3 min read
Source: WIRED
AI Productivity Gains in Fossil Fuels Outweigh Clean Energy Benefits, Study Finds
Photo: WIRED
TL;DR

A new study in npj Climate Action warns that AI’s impact on fossil fuel productivity could increase global energy emissions by 1.2% to 4.8%, far exceeding the emissions from data centers. The research, authored by former Microsoft sustainability employees, argues that AI-enabled extraction and refining create more pollution than AI-driven improvements in renewable energy. This 'enabled emissions' effect is described as a self-reinforcing cycle where tech companies and oil giants collaborate, such as Chevron’s new gas plant for Microsoft data centers in Texas.

Key points

  • Research published in npj Climate Action estimates AI could boost global energy emissions by 1.2% to 4.8%, equivalent to the output of Mexico or Russia.
  • The study, led by former Microsoft employees Will and Holly Alpine, highlights that AI productivity gains in oil and gas outpace benefits in solar and wind sectors.
  • Authors argue that tech companies ignore 'enabled emissions' from fossil fuel production, focusing only on their own operational carbon footprints.
  • Chevron is building a gas plant in Texas to power Microsoft data centers, a deal described by researchers as illustrative of the AI-fossil fuel link.
  • The International Energy Agency estimates AI could increase technically recoverable oil and gas reserves by 5% and cut deepwater project costs by 10%.

Background

This report follows earlier concerns about AI data centers driving fossil fuel use, such as the Texas 'shadow grid' of onsite gas plants reported in July 2026. It also aligns with broader climate trends, including the 2025 State of the Climate report noting record ocean temperatures and the US government’s recent redirection of funds from offshore wind to fossil fuel projects.

How outlets are covering it

WIRED emphasizes the scale of 'enabled emissions' and the self-reinforcing relationship between tech and oil companies, citing the Chevron-Microsoft deal. The Guardian focuses on the quantitative gap, noting that fossil fuel AI adoption is already at scale while renewable AI applications remain largely in pilot stages. The Financial Times headline highlights the core finding that AI boosts oil and gas production more than green energy, though the full article is behind a paywall. All sources agree that current sustainability metrics fail to account for AI’s indirect impact on fossil fuel output.

Why it matters

The findings challenge the narrative that AI is a net positive for climate goals by revealing that its productivity gains in fossil fuels may accelerate emissions far more than its contributions to clean energy. This could undermine climate targets and influence policy debates on AI regulation and fossil fuel subsidies.

What to watch

Researchers and climate analysts will likely push for tech companies to measure and report 'enabled emissions' from fossil fuel production. The study may inform future regulatory frameworks or corporate sustainability standards, while the Chevron-Microsoft deal serves as a case study for the growing integration of AI and fossil fuel infrastructure.

Share this article

Want the full story? Read the original reporting

Read on WIRED