NextFin News - Speaking at the World Economic Forum in Davos on Wednesday, January 21, 2026, Nvidia CEO Jensen Huang delivered a provocative assessment of the global technological landscape, suggesting that the next frontier of artificial intelligence—physical AI and robotics—could allow Europe to surpass the United States in industrial leadership. Huang, appearing alongside BlackRock CEO Larry Fink, argued that while the U.S. dominated the software-driven era of the last two decades, the shift toward "embodied AI" plays directly into Europe’s historical strengths in high-end engineering and manufacturing. However, Huang issued a stern caveat: this transition is entirely dependent on Europe resolving its chronic energy infrastructure and cost issues.
The timing of Huang’s remarks is particularly notable, coming just one day after the inauguration of U.S. President Trump on January 20, 2025. As the new administration in Washington signals a renewed focus on domestic industrial policy and energy independence, Huang’s comments highlight a widening gap in global AI strategies. According to Huang, the world is currently undergoing the "largest infrastructure buildout in history," with over $100 billion in venture capital flowing into AI-native companies in 2025 alone. He described AI as a "five-layer cake" where energy forms the foundational base, followed by chips, cloud services, models, and finally, the application layer. For Europe to succeed, it must secure the bottom layer of that cake.
The core of Huang’s thesis rests on the distinction between digital AI—chatbots and large language models (LLMs)—and physical AI. In the digital realm, the U.S. holds a significant lead through companies like Microsoft, Google, and Meta. Yet, as AI moves into the physical world through humanoid robots and autonomous industrial systems, the competitive advantage shifts toward those who can build the "bodies" for these intelligences. Europe, with its dense ecosystem of automotive, aerospace, and precision machinery industries, is uniquely positioned to lead this "Physical AI" revolution. Huang noted that Nvidia’s Project GR00T, a foundation model for humanoid robots, is already seeing rapid adoption among European robotics startups and industrial giants.
However, the energy crisis remains the primary bottleneck. AI data centers and the factories required to produce advanced robotics are immense power consumers. In 2025, the deployment of Nvidia’s Blackwell architecture pushed global grid demands to record levels. Huang pointed out that while the U.S. has leveraged its domestic energy resources to lower the cost of compute, Europe continues to face higher electricity prices and regulatory hurdles that could stifle the buildout of the necessary "AI Factories." Without a radical overhaul of its energy policy to provide inexpensive, reliable power, Europe risks squandering its industrial advantage to more energy-efficient competitors.
From a financial perspective, the stakes are unprecedented. Nvidia itself reached a $5 trillion market capitalization in late 2025, reflecting its role as the "toll booth" for this new economy. Huang’s forward-looking analysis suggests that the market is shifting from "training" models to "inference"—the actual application of AI in real-time environments. This shift is expected to drive a massive wave of capital expenditure. For instance, TSMC has already announced plans to increase its investment to over $50 billion in 2026 to meet the demand for AI chips. If Europe can align its energy infrastructure with its industrial capabilities, it could capture a significant share of this value creation, which Huang described as a "once-in-a-generation" opportunity.
Looking ahead, the geopolitical implications are equally profound. As U.S. President Trump begins his term with a likely emphasis on "America First" technological dominance, Europe’s ability to carve out a leadership role in robotics will be a test of its strategic autonomy. Huang’s message to European leaders was clear: treat AI infrastructure as a national priority on par with roads and electricity. By leveraging open models and local data, European nations can develop "national intelligence" tailored to their specific industrial needs. The window for this transition is narrow, but if the energy puzzle is solved, the 2026-2030 era could see a rebalancing of global tech power in Europe’s favor.
Explore more exclusive insights at nextfin.ai.
