Surge in AI Data Center Energy Needs Could Fuel Natural Gas Demand
The energy landscape is poised for a significant shift as artificial intelligence data centers are projected to drive a substantial increase in power usage. Investment bankers at Tudor Pickering Holt & Co have highlighted a potential surge in natural gas demand as a direct consequence of this growth. Their analysis suggests an additional 8.5 billion cubic feet per day (bcfd) of natural gas may be necessary to accommodate the burgeoning energy requirements.
In the face of escalating power needs, particularly from cutting-edge Generative AI technologies, US power and technology companies are raising alarms over the pace at which the country’s electrical infrastructure is expanding. This concern has led some data centers to seek alternative solutions, including forging direct agreements with power producers or investing in their own energy resources.
The increasing demand for electricity is contributing to a backlog of power generation and energy storage project requests. Data from Lawrence Berkeley National Laboratory indicates that the queue for grid connections has ballooned to 2,600 gigawatts in 2023, up from 2,000 gigawatts the previous year.
Amidst fluctuating natural gas prices, which hit a three-and-a-half-year low in February at $1.61 per million British thermal units (mmbtu), the report anticipates an average price of $4 per mmbtu during the latter half of the decade. The mild winter weather played a role in the recent price dip, prompting producers to scale back on production.
Pipeline operators such as Kinder Morgan, Williams, and Energy Transfer are expected to be well-positioned to capitalize on the increased gas demand. Gas producers like EQT and Chesapeake Energy are also predicted to see benefits from this trend.
The current power demand from data centers stands at 11 gigawatts (GW), with projections estimating growth to 42 GW by 2030. To support this expansion, the report forecasts that approximately 2.7 bcfd of incremental natural gas will be needed by the end of the decade.
The interplay between technology’s insatiable appetite for energy and the resources required to sustain it continues to shape the future of both the tech industry and energy markets. As AI continues to evolve and integrate into various sectors, its impact on power infrastructure and natural resource consumption will likely remain a critical topic for stakeholders across both fields.
