OpenAI outlines strategy to curb data-center energy costs as AI power demand surges

OpenAI on Tuesday announced a new Stargate Community plan designed to ensure that its massive expansion of AI data centers does not push up electricity prices for local communities. The initiative comes as access to affordable and reliable energy is emerging as one of the biggest constraints on the growth of artificial intelligence.

The plan is part of OpenAI’s broader Stargate project, a $500 billion, multi-year effort to build large-scale AI data centers for model training and inference. Stargate is backed by major partners including Oracle, and received public support from Donald Trump when it was first unveiled in January 2025.

Under the new framework, each Stargate data-center site will adopt a locally tailored community energy plan shaped by input from residents, utilities and regional authorities. OpenAI said the goal is to “pay its way on energy,” rather than drawing on existing grids in a manner that could raise costs for households and businesses.

Depending on location, OpenAI said the approach could include fully funding dedicated power and energy-storage infrastructure, or paying for new electricity generation and transmission capacity to be added to the local grid. In all cases, the company aims to ensure that incremental demand from its data centers is matched by incremental supply that it helps finance.

The announcement highlights how energy has become a strategic bottleneck for the AI industry. As models grow larger and data centers multiply, power consumption is rising sharply, forcing technology companies to look beyond traditional cloud procurement and toward direct involvement in energy infrastructure.

OpenAI’s move follows a similar push by Microsoft, which last week outlined plans to reduce water usage at its U.S. data centers and mitigate any knock-on effects on local electricity prices. Microsoft has said it will pay utility rates that fully reflect its power needs and work with local providers to expand generation and grid capacity where required.

Together, these initiatives signal a broader shift in how major AI developers approach sustainability and community impact. Rather than relying solely on utilities to absorb surging demand, leading firms are increasingly positioning themselves as long-term partners in building the power systems needed to support the next phase of AI growth.