Elon Musk warns AI could face 15 gigawatt power shortfall by 2027

Power shortage as a bottleneck for AI

Elon Musk has warned that AI could face a significant electricity shortage as soon as 2027, according to YourStory. He stated that the rapid expansion of AI chip production is outpacing the growth of electricity infrastructure, which could create a 15 gigawatt shortfall for AI chips by that year. The Tesla and SpaceX CEO described the issue as "not the distant future" during a virtual address at a G20 session on emerging technologies.

Disparity in growth rates

Musk highlighted a stark contrast in growth rates: AI chip production is increasing by 40% to 50% annually, while electricity availability outside China is growing at only 10% to 20% per year. This gap suggests that electricity could soon become a bigger constraint for AI companies than access to processors, which has been the primary focus until now.

Data centres as a critical constraint

The surge in AI chip production necessitates more data centres, which require substantial electricity for both computing and cooling. According to Musk, building this infrastructure takes significantly longer than producing chips, adding another layer of complexity to AI expansion. Meta CEO Mark Zuckerberg echoed these concerns at the same G20 session, noting that constructing data centres demands hundreds of thousands, if not millions, of skilled trade workers.

Zuckerberg added that Meta is already struggling to hire enough workers to build the facilities it needs. This labor shortage compounds the challenges of securing land, grid connections, permits, and cooling systems, all of which are now critical to AI growth.

Community and regulatory challenges

The push for new data centres is already creating tensions in some communities. According to YourStory, these facilities have raised concerns over rising electricity prices, noise pollution, and strain on local infrastructure. Regulatory approvals and community pushback could further delay or limit the expansion of AI infrastructure.

Energy infrastructure as a competitive advantage

Musk argued that countries capable of rapidly expanding their power generation could gain a competitive edge in attracting AI investments. Large data centres can drive economic benefits like tax revenue, but only if governments can provide reliable and scalable electricity. This shift could turn energy infrastructure into a key factor in the global AI race, alongside advancements in chips and models.

He also projected that AI could add 20% to 30% to the global economy, equivalent to $20 trillion to $30 trillion annually, alongside productivity gains in software and engineering. However, these projections hinge on overcoming infrastructure challenges.

What this means

Lazyfounder analysis — our interpretation, not reported fact.

Musk’s warning shifts the conversation from hardware scarcity to infrastructure bottlenecks. For founders, this means AI scalability is no longer just about securing chips or talent—it’s about power availability. Startups in data-centric fields like AI, cloud computing, or even crypto may need to factor electricity costs and grid reliability into location decisions. Investors, meanwhile, might start favoring regions with strong energy infrastructure or innovative power solutions, such as nuclear or renewable microgrids. The takeaway: energy could become the next moat for AI companies, and those who plan for it early may avoid costly disruptions.

Key takeaways

• AI companies may face electricity shortages before chip shortages, reshaping expansion strategies.
• Data centre construction is now a critical bottleneck due to labor shortages and regulatory hurdles.
• Power infrastructure could determine which countries and regions attract AI investments.
• Founders should assess electricity availability and costs when planning data centre or compute-heavy operations.