AI energy demand is skyrocketing, and tech giants admit that natural gas is among the options.

As the artificial intelligence industry rapidly expands, representatives from the U.S. technology and energy sectors gathered in Oklahoma City this week to discuss meeting the soaring energy demands of AI data centers. At a conference organized by the Hamm Institute for American Energy, executives from Amazon and Nvidia publicly stated for the first time that all energy options, including natural gas and other fossil fuels, are being considered to meet the short-term surge in computing power demand.

Kevin Miller, Amazon's Vice President of Global Data Centers, highlighted during a panel discussion: "To secure the energy needed for the grid, we will need to adopt an 'all-encompassing' energy strategy for a period of time." He emphasized that while Amazon's long-term goal remains achieving net-zero carbon emissions, the most immediate task is ensuring ample computing resources for customers.

Miller added that Amazon remains the world's largest buyer of renewable energy and is forextrustindex in advanced nuclear and carbon capture technologies. However, these green technologies are not expected to be deployed on a large scale until the 2030s, and the company urgently needs stable and reliable power sources to support business growth. He stressed: "Our target is net-zero emissions by 2040, but in the short term, securing sufficient power is crucial."

Nvidia expressed a similar stance. Josh Parker, the company's Senior Director of Corporate Sustainability, stated that while Nvidia highly values environmental impact, "all options are on the table" given the current energy constraints. Parker noted that different customers prioritize clean energy to varying degrees, but "regardless, at the core, we need electricity."

Jack Clark, co-founder of AI company Anthropic, further highlighted the real challenges. He predicted that by 2027, the additional power demand from the AI sector would reach 50 gigawatts, equivalent to the generating capacity of 50 nuclear reactors. Clark pointed out that while AI's long-term needs are likely to spur the development of new energy technologies, the industry must pragmatically face the existing energy structure in the short term.

However, the atmosphere became subtle when discussing whether coal could potentially power AI. Despite recent executive orders signed by U.S. President Trump aiming to revive coal production through AI, Amazon and Nvidia executives were cautious in their stance on coal during the panel, refraining from outright support.

Clark explicitly stated: "There are broader choices beyond coal. We will certainly consider it, but it is by no means a priority." This statement reflects that although there might be greater openness to energy sources in the short term, the idea of coal as a pillar of energy in the AI era remains difficult for the tech industry to widely accept.

As AI's energy demands rapidly increase, the collaboration and conflicts between the tech and energy sectors concerning power sources are also intensifying. In the future, balancing the need for high-performance computing with environmental responsibility will become a significant challenge for the U.S. and globally.

The market carries risks, and investment should be cautious. This article does not constitute personal investment advice and has not taken into account individual users' specific investment goals, financial situations, or needs. Users should consider whether any opinions, viewpoints, or conclusions in this article are suitable for their particular circumstances. Investing based on this is at one's own responsibility.

Very good!（5628）