CITIC Securities: Computing Power Drives Power Restructuring, US Self-Supply Opens Trillion-Dollar New Track

robot
Abstract generation in progress

China Securities Journal Research Report states that the explosive growth in AI computing power is causing a disruptive reshaping of the global electricity supply and demand landscape. As the core hub of global AIDC construction, the United States is facing an unprecedented electricity shortage crisis. The severe mismatch between the grid construction cycle (3-8 years) and AI server deployment cycle (6-12 months), combined with the structural flaws of the U.S. grid’s natural disconnection and insufficient cross-regional transmission capacity, is driving autonomous power supply for AIDC from an “optional option” to a “rigid necessity.” In March 2026, seven major U.S. AI giants signed the “Electricity Cost Payer Protection Commitment,” explicitly stating “full self-sufficiency for new electricity use and full bearing of costs,” officially marking the beginning of the industrialization year for autonomous power supply in AIDC.

The full text is as follows

Electricity|Power-driven Reshaping of Electricity, U.S. Autonomous Power Supply Opens Trillion-Dollar New Track

The explosive growth in AI computing power is causing a disruptive reshaping of the global electricity supply and demand landscape. As the core hub of global AIDC construction, the United States is facing an unprecedented electricity shortage crisis. The severe mismatch between the grid construction cycle (3-8 years) and AI server deployment cycle (6-12 months), combined with the structural flaws of the U.S. grid’s natural disconnection and insufficient cross-regional transmission capacity, is driving autonomous power supply for AIDC from an “optional option” to a “rigid necessity.” In March 2026, seven major U.S. AI giants signed the “Electricity Cost Payer Protection Commitment,” explicitly stating “full self-sufficiency for new electricity use and full bearing of costs,” officially marking the beginning of the industrialization year for autonomous power supply in AIDC.

Triple core contradictions stacking, AIDC autonomous power supply upgrading from an optional solution to an industry rigid standard.

The U.S. power system supply continues to weaken. We expect the share of general base load power sources to drop to 59.9% by 2030, with surplus electricity significantly declining from 2024 onwards. The power capacity of single AIDC cabinets has surged to 30-100kW, becoming a key driver of increased electricity demand; combined with the structural flaws of the U.S. grid’s natural disconnection, insufficient cross-regional transmission, and queue times exceeding 7 years for grid connection, as well as the “full self-sufficiency for new electricity use” policy signed by the seven AI giants in March 2026, these factors collectively push autonomous power supply from a spontaneous corporate behavior to a mandatory industry requirement.

U.S. AIDC clusters show regional concentration, with resource endowments giving rise to three differentiated autonomous power supply models.

U.S. AIDC’s total IT capacity is highly concentrated in three core regions. Texas relies on cheap natural gas and an independent grid to build a “off-grid” intelligent computing park mainly powered by gas turbines with photovoltaic energy storage as a supplement; Arizona leverages its sunlight advantage to explore a 24/7 zero-carbon energy solution combining photovoltaics and long-term storage; Northern Virginia, due to grid saturation and environmental regulation pressures, shifts toward clean power paths mainly using nuclear power direct supply and SOFC fuel cells.

Eight technological routes evolving in gradients, with global market space expected to surpass $1.2 trillion from 2026-2030.

A clear technological echelon is forming for global AIDC autonomous power supply, with a power source structure trending toward “gas transition, diversified low-carbon.” Existing projects are mainly based on traditional grids and diesel generation; ongoing projects are fully taken over by gas-fired power matrices; planned projects are inclined toward low-carbon technologies such as photovoltaic storage, SOFC, SMR, and geothermal. We estimate that gas turbines account for 43% of the total market space, energy storage 24%, and photovoltaics, SOFC, SMR, and geothermal combined account for 33%.

Gas turbines become the short-term core solution, with global capacity bottlenecks creating spillover opportunities for Chinese industry chains.

Driven by AIDC demand, global orders for gas turbines are exploding. Leading companies like GEV and Siemens Energy have order delivery ratios over three times higher, with production extending into 2029, and capacity-demand gaps continuing to widen. To ease delivery pressures, overseas leaders are accelerating the transfer of non-core and some core component capacities to China. Chinese manufacturers with the ability for mass production of hot-end components are expected to benefit deeply. Additionally, increased demand for aircraft engine modifications and higher-pressure diesel units also present structural growth opportunities.

Multiple points of effort in energy storage, SOFC, SMR, and geothermal to build a low-carbon power supply system for the short, medium, and long term.

Energy storage benefits from dual demand on the load side and power source side in AIDC. We project industry growth rates reaching 158% by 2026, with Chinese manufacturers leveraging cost advantages to accelerate overseas expansion. As the fastest-deploying low-carbon solution, SOFC can be delivered within 90 days and reduces carbon emissions by over 30%, with costs continuously decreasing and scaling orders arriving. SMR small modular reactors and enhanced geothermal systems, as long-term low-carbon power sources, are advancing multiple technological routes in parallel. We expect these to gradually commercialize after 2028, opening long-term growth space.

Risk factors:

  1. Risk of AIDC demand falling short of expectations: If AI technology development slows or commercialization falls short, AIDC construction progress and electricity demand will be below forecasts, impacting autonomous power equipment demand.
  2. Risk of policy support falling short: If U.S. state policies supporting AIDC autonomous power supply are adjusted or environmental regulations tighten, leading to project approval delays, industry development pace will be affected.
  3. Risk of lagging core technological breakthroughs: If commercialization of SOFC, SMR, EGS, and other technologies is slower than expected, the replacement speed of low-carbon power sources will be impacted.
  4. Supply chain risks: Disruptions in core components of gas turbines could delay equipment delivery and increase costs.
  5. Geopolitical and trade barrier risks: If the U.S. implements trade restrictions targeting Chinese power equipment, it will affect Chinese companies’ overseas expansion.

▍Investment strategies:

It is recommended to focus on the following core lines:

  1. Gas turbine industry chain: Prioritize layout of key component suppliers already integrated into global leading supply chains and capable of mass production of hot-end parts.
  2. Energy storage segment: Focus on system integrators and core equipment suppliers benefiting from the explosive demand for dual storage on the load and power source sides in AIDC.
  3. SOFC: Pay attention to fuel cell leaders that have achieved commercialization and are experiencing rapid order growth.
  4. SMR nuclear power: Invest in core equipment and material suppliers, and seize long-term technological iteration opportunities.

(Source: First Financial)

View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • Comment
  • Repost
  • Share
Comment
Add a comment
Add a comment
No comments
  • Pin