Global trading platform is full of "AI bubbles"! Even France and Germany, which are lagging behind in AI competitions, cannot stay out of it.

If the artificial intelligence investment frenzy is indeed a historical investment bubble similar to the 2000 Internet bubble, then it has already inflated the overall market value of many "non-pure AI computing hardware target companies" outside of the "American tech giants" such as NVIDIA Corporation, Alphabet Inc. Class C, and Broadcom Inc., to the point of being bloated. Countries such as India and many European countries that are seriously lagging in the global "AI competition wave" have also been brought into the discussion of the stock market closely related to the AI bubble by financial market traders. This global stock market-sweeping "AI bubble rhetoric" also implies that even if a country's stock market does not have pure AI computing hardware targets like the American chip giants NVIDIA Corporation, Broadcom Inc., Micron Technology, Inc., and AMD, as well as lacking super cloud computing giants like Alphabet Inc. Class C, Microsoft Corporation, and Amazon.com. Inc. that directly benefit from the surge in cloud services and cloud-based AI computing power, it is still being caught up in the bubble discussion due to the large-scale market speculation in data center operators, electrical and core power equipment manufacturers, and other related data center targets. In the case of the Indian stock market, some Wall Street traders claim that the country's stock market has already fallen into an AI bubble narrative, as it severely lacks pure AI computing leaders like NVIDIA Corporation, Broadcom Inc., and Micron. However, investors are turning to local data center support companies from small cloud infrastructure providers, data center operators, liquid cooling technology providers, power equipment manufacturers, to power generation companies which are expected to benefit significantly from the construction of supporting infrastructure for AI data centers, making them one of the hottest equity trades globally. These data center support companies have generally seen gains of over 100% and their valuations have become exorbitantly high. Compared to the two AI superpowers, China and the U.S., as well as Japan, South Korea, and India, which have been investing heavily in AI in recent years, Europe may be a 100% laggard in the AI field. However, some of the long-standing traditional industrial giants in France and Germany have jumped on this unprecedented wave of artificial intelligence super bull market, and investors who have long bet on these stocks are just as excited since the bull market in the U.S. since 2023 at least until the slight global stock market correction earlier this week. The AI bubble is also sweeping through the French and German stock markets The race towards the "divine virtual" AI competition has hit bottlenecks in land, labor, and electricity the basic human technological scale. The traditional industrial giants in France and Germany, including Legrand SA and Schneider Electric SE (both founded in the 19th century), are eagerly providing "picks and shovels" for this AI data center gold rush. Legrand's stock price has even outperformed that of the "AI chip superpower" NVIDIA Corporation since the beginning of this year. For traders, AI data centers' massive NVL72 cabinet power equipment, heat dissipation equipment, power management tools, and all other key infrastructure necessary for the projected surge in AI infrastructure spending up to 2030 to potentially reach $7 trillion are crucial. With the growing demand for high-energy AI data centers driven by AI chips and other AI computing infrastructure, electricity supply becomes essential, giving rise to the market view that "AI ends in electricity." An International Energy Agency (IEA) forecast report predicts that global data centers' electricity demand will more than double by 2030, reaching approximately 945 terawatt-hours (TWh), slightly higher than Japan's current total electricity consumption. AI applications will be the most important driver of this growth, and by 2030, the overall electricity demand for AI-focused data centers is expected to more than quadruple. As the AI computing wave brings about massive power distribution requirements, Schneider Electric, one of the world's largest electrical equipment manufacturers based in France, is reaping the benefits of this "AI dividend." In the second quarter, the company achieved rare double-digit growth in revenue related to AI data centers. Schneider Electric's core power equipment supply for AI data centers, which include medium/low-voltage power distribution, UPS, battery storage, HVDC busbars, liquid cooling, and DCIM software, are essential to companies like Meta, Microsoft Corporation, Amazon.com, Inc., and Alphabet Inc. Class C which operate super large-scale AI data centers. The continuously increasing scale of AI training/inference clusters has made "power + cooling" a core underlying resource that determines computing boundaries. Schneider is benefiting directly from this super AI infrastructure wave with its full-chain power system solutions from medium voltage to rack endpoints, software and hardware digital twins, and liquid cooling mergers and acquisitions. The urgent need for liquid cooling systems in super large-scale AI data centers has made it a focus area for Schneider in recent years, and in 2024, Schneider acquired Motivair to enter the submerged/direct liquid cooling plate and Cooling Distribution Unit (CDU) markets. These seemingly old-school power giants are not as glamorous as fabless light asset chip giants like NVIDIA Corporation and Broadcom Inc., but they are crucial for training large language models and handling massive AI inference workloads according to a research report by Bloomberg Intelligence's senior analyst Omid Vaziri, the energy consumed by these models may be more than 10 times that of traditional CPU-level data centers. This is an important part of what venture capital firm Air Street Capital calls the AI "industrial era": gigantic projects akin to those in the era of the pharaohs are being rapidly rolled out, with power supply becoming a primary constraint. According to Dell'Oro analyst Alex Cordovil, the AI data center infrastructure market grew by 18% in the second quarter of this year, reaching $8.9 billion, and he expects this strong growth momentum to continue until 2026. European industrial companies are now littered with mentions of the term AI in their earnings calls. Siemens Energy AG's supply capacity of core power equipment such as turbines for global gas power plants is becoming tight due to the massive power demand brought about by AI training/inference. Can this AI investment frenzy in Europe continue? But how long can this good situation in the European market last? Senior analyst Bill Ford from the investment firm General Atlantic warns that there is currently an "irrational exuberance" in all markets, and the strong profit performance of large tech companies has not dispelled fears of a long-term adjustment after the bubble burst. The current enterprise value of Legrand and ABB is about 20 times their underlying profits, hitting historical highs such high valuations are dizzying for this once relatively ignored corner of the stock market. Data center DRIVE the expected capital expenditure distribution globally until 2030 far exceeds NVIDIA Corporation/AMD AI chips themselves Risk is abundant in the European region. AI spending could be cut due to diminishing returns, and the ability of European suppliers to maintain a competitive advantage may weaken. Giant, energy-hungry data centers may face opposition from local communities. While their potential economic value is still uncertain, the impact on electricity bills could be significant, and if the giants who operate these data centers notice a sharp rise in power costs, they may immediately halt or scale back their data center expansion plans. Once the global tech stock market collectively crashes, the ensuing aftermath cannot be sugar-coated. However, European industrial companies still have some safety nets. The significant slowing down of new data center construction should not diminish the need for renovations and upgrades to existing massive data centers in Europe. Investments in these vast parks may even shift from the U.S. to Europe; Europe's infrastructure projects under construction amount to $183 billion, leading East Asia's approximately $79 billion. Additionally, these companies are not entirely tied to the AI data center frenzy like NVIDIA Corporation and have wider connections to real estate and electrification-related businesses. While the European real economy has not experienced a robust recovery, when trying to hedge against the rise of the far-right by revitalizing traditional manufacturing growth, European leaders still have enough room to increase infrastructure investments, all it takes is political will. Equally important is whether European companies can adeptly adopt groundbreaking AI applications like AI intelligent agents. EU data shows that only 13.5% of companies used tools related to generative AI last year. Increasing this number in a region struggling with declining birth rates and low productivity could completely change the game. Large-scale investments in telecommunications and technology have helped drive France's stunning economic growth in the third quarter. During the oil shock of the 1970s, there was a popular saying: "France has no oil, but it has creativity." Europe still lacks this black liquid, but it has enough data and data center power infrastructure. The future of these European continental countries may depend on how well they can use this data. In the era dominated by ChatGPT, selling picks and shovels is certainly a decent business, but if one can also dig up gold along the way, then that's not a bad thing.