Power Shortages Hinder Artificial Intelligence Expansion
Goldman Sachs warns that grid infrastructure limitations are creating a bottleneck for the rapidly growing AI industry.
Updated on Oct. 20, 2026 in Data Centers
Live Poll
Is the current level of debt-fueled investment in AI infrastructure a threat to national economic stability?
Goldman Sachs reports that power infrastructure constraints are slowing the development of artificial intelligence, with the U.S. accounting for 60% of projected global power demand growth. The report highlights that physical infrastructure development is currently failing to keep pace with the underlying technology.
Why it matters
The mismatch between rapid AI growth and the availability of electricity puts significant pressure on power grids and increases reliance on heavy corporate debt. If utilities cannot scale generation, the growth of the AI ecosystem faces long-term operational delays.
Global data center power demand is projected to rise 170% by 2030, with U.S. infrastructure bottlenecks leaving 2,600 gigawatts of capacity in limbo. Projects now face a median five-year wait time for grid connections, with 86 projects delayed or canceled since January 2026.
The players
Goldman Sachs
Goldman Sachs is a leading global investment banking, securities, and investment management firm that provides a wide range of financial services.
The details
Tech giants are increasingly exploring plans to fund power generation and transmission lines directly, as they currently account for 40% of ecosystem credit fundraising. This expansion is being financed significantly through corporate debt, which is expected to cover 35% of hyperscaler capital expenditures by 2027.
Timeline
January 2026 marked the beginning of the tracking period for project delays and cancellations.
Data center power demand is projected to increase 170% between 2025 and 2030.
Utility companies have planned $1.3 trillion in capital investments for 2026 through 2030.
Global capital spending for data and power infrastructure will total $7.6 trillion from 2026 to 2031.
Hyperscalers are projected to issue $420 billion in debt during 2027.
The Tech Race
This development highlights the shift from software-focused competition to a race for physical energy dominance. It marks a departure from purely digital innovation as tech giants increasingly become primary financiers of national electrical infrastructure.
As tech companies rush to secure power, users may experience higher service costs as companies pass on capital expenditures to customers. The ongoing strain on the grid also introduces potential uncertainty regarding regional energy reliability as utility companies prioritize hyperscale projects.
The takeaway
The rapid acceleration of AI is forcing tech companies to become utility-scale energy investors to survive. Readers should note that the physical constraints on power generation are now as significant to the tech sector's success as software breakthroughs.
Further reading
For more context on the energy needs of modern computing, visit the Data Centers section.
Live Poll
Is the current level of debt-fueled investment in AI infrastructure a threat to national economic stability?
