"The "new mainstream" under the AI boom! Power grid technology stocks lead the new craze of investment in AI infrastructure."

With the prosperity and expansion of AI data centers and the acceleration of global electrification, the stock prices of those technology companies that provide services to the power grid system are rapidly soaring. Wall Street financial giants such as J.P. Morgan are predicting that the bull market trend of these companies will continue long-term in the unprecedented wave of AI and electrification. Although some corners of the energy market may seem "bubbly" recently, there is one sector that Wall Street is betting will not be trapped in any meaningful bubble: the so-called "Grid Tech Stocks". The "AI bubble theory" that had a significant impact on global stock markets in November has been refuted by many on Wall Street recently. Top analysts from companies such as Morgan Stanley, Citigroup, Loop Capital, and Wedbush believe that the global investment wave in AI infrastructure, centered around AI computing power hardware, is far from over. They predict that this wave, driven by the unprecedented demand for AI reasoning computing power, will continue until 2030, with an estimated investment scale of $3 to $4 trillion. According to Dan Greenhaus, Chief Strategist at Solas Alternative Asset Management, the brief "AI bubble panic" that occurred earlier in November is likely to have ended. He does not expect significant changes in market dynamics as we move towards the new bull market of 2026, as both the fundamentals of the AI investment theme and the potential rate cuts by the Fed are favorable for investors. The "Grid Tech Stock frenzy," born out of the AI investment theme, surpasses the AI infrastructure investment craze Steve Tusa, Managing Director and Senior Equity Analyst at JPMorgan Chase & Co., stated that despite a 30% overall increase in the grid tech stock sector this year, grid tech stocks remain attractive investments within the global stock market. Grid tech stocks encompass a range of well-known hardware manufacturers, software developers, and utility-scale battery product installers. Tusa advises investors to take advantage of minor price corrections. "At this stage, any corrections are buying opportunities." Tusa from J.P. Morgan added. "Grid Tech Stocks" refer not to traditional utility companies, but to a group of "infrastructure technology + equipment manufacturing" companies that provide equipment, software, and engineering services to the power grid. They ride the wave of AI, data centers, electric vehicles, and renewable energy without being solely reliant on the popular AI race. In the current market narrative, they are seen as beneficiaries of a combination of factors such as "rising electricity usage in AI data centers + electrification (EVs, heat pumps, industrial electricity) + renewable energy integration + large-scale overhaul of aging power grids." Companies like Vertiv Holdings, a leader in liquid cooling technology, providing microgrid and energy storage solutions to large AI data centers led by Microsoft Corporation, Alphabet Inc., and Meta, have seen their stock prices rise by about 60% this year. Tusa stated that despite the significant premium at which the company's stock trades compared to the S&P 500 index, its strong growth "proves" that this premium is justified. Other types of grid tech stocks have also seen significant gains this year, echoing the prosperity of data centers. South Korean transformer manufacturer Hyosung Heavy Industries Corp. and LS Electric Co. have seen their stocks soar by about 400% and 230% respectively. In the U.S. market, inverter system manufacturer SolarEdge Technologies Inc. has doubled its stock price, while power engineering company Willdan Group Inc. trades near its historic highs, with an increase of nearly 200% year to date. As the scale of AI training/inference clusters continues to grow rapidly, "power + heat dissipation" has become a core underlying resource in determining the limits of computing power. Schneider Electric, one of the largest global manufacturers of electrical equipment based in France, has benefitted from the "AI dividend". In the second quarter, the company achieved rare double-digit growth in endogenous revenue related to AI data centers. Schneider Electric's long-standing focus on medium/low voltage distribution, UPS, battery storage, HVDC busbars, liquid cooling, and DCIM software is essential for large-scale AI data centers such as Meta, Microsoft Corporation, Amazon.com, Inc., and Alphabet Inc. Class C. The rapidly growing AI training/inference cluster size has made "power + heat dissipation" the core underlying resource determining the computing power boundaries. Schneider has positioned itself as a direct beneficiary in this large-scale AI infrastructure wave with solutions covering the entire chain from medium voltage to rack-end power systems, software/hardware digital twins, liquid cooling, and acquisitions. The urgent need for liquid cooling systems in large-scale AI data centers has been a focus area for Schneider in recent years. In 2024, Schneider acquired Motivair, entering the immersion/direct liquid cooling plate and cooling distribution unit (CDU) market. "This is not just about AI," commented Tim Chan, Director of Sustainable Development Research at Morgan Stanley in the Asia-Pacific region, excluding Japan, "Overall energy demand is growing rapidly." Fidelity International's core view is that there is currently "a very long structural change" underway. Gabriel-Wilson Otto, Director of Sustainable Investment Strategy at Fidelity, stated that factors other than AI are still playing a "larger role" in driving rising energy demand in developing economies, which should support grid tech stocks globally. With climate change leading to longer periods of extreme weather, developed and developing countries are in urgent need of upgrading their aging power grids. According to a recent research report by BloombergNEF, total global spending on the power grid is expected to increase by 16% this year, to $479 billion, and is projected to rise to $577 billion by 2027. The International Energy Agency (IEA) predicts that by the end of this decade (2030), the electricity demand of data centers (including AI and non-AI data centers) will at least double, with almost every new large power plant needing to be connected to the grid system. As shown in the chart above, global investment in power grids has surged in recent years, with large-scale integration of electrification, AI, and renewable energy systems driving global grid construction. The Nasdaq OMX Clean Edge Smart Grid Infrastructure Index tracks companies related to the power grid infrastructure business. This year, it has risen by about 30%, outperforming other major U.S. stock indices. The Nasdaq 100 index, which includes technology giants such as NVIDIA Corporation, Apple Inc., and Microsoft Corporation, has risen by about 22% during this period. The grid index currently trades at 21 times forward earnings, making it undervalued compared to the Nasdaq 100 index. Of course, when global stock markets were severely hit last month due to concerns about an AI bubble bursting, grid tech stocks also experienced significant declines. This highlights the fact that not all investors are convinced that this sector can thrive amidst potential slowdowns in AI spending or AI monetization pathways. "The grid tech stock investment theme will continue to be a structural winner until the end of 2026," said Lisa Audet, Founder and Chief Investment Officer of Tall Trees Capital Management LP, a boutique hedge fund focused on energy transition. However, she also pointed out that many positive news has already been reflected in the strong stock price increases this year. She cautioned that investors must currently be "very discerning in terms of valuations and cyclical factors." Many grid upgrade activities may require deep and long-term collaboration with utility companies, or at least data provided by these regulated monopoly corporations, which could slow down or impede investments. As customer bills increase, some U.S. states are increasing scrutiny, potentially hindering deployments in certain areas where grid technology is seen as too risky. Different utility companies, states, regional grid operators, and broader regulatory structures will lead to varying adoption rates. Hedge funds reporting positions to Hazeltree, a U.S. data provider, are still net long towards the Nasdaq grid index. Data shows that as of the end of September, 66% of companies in the index had more long positions than short positions, up from 59% a month earlier. Hazeltree tracks 108 of the 113 component stocks in the index, with about 600 funds of various sizes voluntarily reporting their holdings to the platform. Maybe a decades-long super investment cycle "Grid infrastructure is not purely about AI; you can see it as a stock market narrative of 'the chicken finally comes home to roost'," said Garvin Jabusch, Chief Investment Officer at Green Alpha Advisors. Since 2023, the Nasdaq grid tech index has risen for three consecutive years, although the previous gains were more moderate. As the AI hype has brought grid infrastructure into the spotlight of the global stock market, "the market is finally pricing in the positive growth that should have been obvious a long time ago," Jabusch said in a report. This is particularly evident in the U.S. and Europe, where most grid systems were built decades ago when power was generated solely by coal-fired power plants and transmitted unidirectionally from utility companies to consumption markets. Today, renewable energy is rapidly rising, and home storage batteries paired with rooftop solar panels can feed a substantial amount of power back to the grid, necessitating upgrades from transformers to transmission lines for a modernized system. Venture capitalists also see significant opportunities in this area. "We don't need to bet on data centers becoming the core growth engine; growth in other areas is also very significant," said Evan Caron, Chief Investment Officer at Montauk Capital, which supports early-stage energy and grid startups. He believes that the hot trend of building large-scale AI data centers globally is like "throwing gasoline on a fire that is already burning." Alex Darden, Head of Infrastructure Investments for the Americas at EQT Partners Inc, a global private equity firm, stated in an interview that despite some degree of hype and fear of a bubble surrounding AI, the continued and growing positive factors, coupled with historical data showing inadequate long-term investment in grid infrastructure, are creating "significant investment opportunities." "And this is not just a one-year opportunity in 2026," Darden said in the interview. "What we are entering now is a 'super investment cycle' for power infrastructure that may last for years, even decades." The essence of the global AI competition is a competition for AI computing power infrastructure, and the core foundation driving AI clusters is a stable and massive power supply system. Therefore, the demand for electricity in AI data centers is rising at an unprecedented rate, with AI now embodying "power consumption." The highly energy-demanding AI data centers, whose scale is expanding exponentially due to the demand for AI computing power, cannot be met even by the continuous expansion of large data centers in recent years. After Alphabet Inc. Class C released the Gemini3 AI application ecosystem in late November, this cutting-edge AI software quickly gained worldwide popularity, driving a sudden increase in AI computing power demand for Alphabet Inc. Class C. The Gemini3 product line brought a massive processing volume of AI tokens upon its release, prompting Alphabet Inc. Class C to significantly reduce free access to Gemini 3 Pro and Nano Banana Pro products and impose temporary restrictions on Pro subscribers. The recent trade export data in South Korea showing strong demand for HBM storage systems and enterprise-grade SSDs further validates the Wall Street shout of "the AI boom is still in the early stages of construction of AI computing power infrastructure". A forecast report from the International Energy Agency (IEA) shows that by 2030, global data centers' electricity demand will more than double, reaching about 945 terawatt-hours (TWh), slightly higher than Japan's current total electricity consumption. AI applications will be the main driving force behind this growth; by 2030, the overall electricity demand of data centers focused on artificial intelligence will at least quadruple.