Two major developments in the U.S. tech industry highlight how quickly the artificial-intelligence (AI) race is shifting power dynamics: one involves Alphabet Inc. (the parent of Google), and the other Nvidia Corporation.
On one hand, Alphabet is on track to hit a $4 trillion market valuation, driven by its renewed focus on AI tools and infrastructure. Having gained almost 70 % so far in 2025, its shares and business momentum reflect a strong investor endorsement of its cloud and AI-model ambitions.
On the other hand, Nvidia, long seen as the dominant supplier of the high-performance chips that power AI workloads, is seeing pressure. Reports that Meta Platforms, Inc. is in talks to use Google’s custom AI chips (tensor processing units or TPUs) in its data centres from 2027 have sent Nvidia shares notably lower. This suggests a potential pivot in how major AI players procure hardware, and signals that Nvidia’s dominance may face a meaningful challenge ahead.
Why this matters for investors
• Valuations are being re-rated in real time based on AI strategy execution, not just legacy business strength. Alphabet’s leap reflects that.
• Supply-chain and hardware shifts (i.e., who builds the chips) may create winners and losers amongst large-cap tech stocks, making portfolio differentiation important.
• The convergence of cloud, AI-model development, and custom hardware is increasingly where value is being built, and where risk is accumulating (e.g., about rising valuations, competitive disruption, and regulatory scrutiny).
Key take-aways
• Alphabet’s AI-driven momentum reinstates it among the very largest companies globally, reaching the $4 trillion threshold would place it alongside only a handful of peers.
• Nvidia’s slip is not about a failing business; its fundamentals remain strong, but more about market perception of future growth and competitive threat from within the tech ecosystem.
• For investors with large tech exposures (or those looking at tech-heavy portfolios), this moment underscores two things: the importance of timing (when the market pivots) and the importance of differentiating between “what has been” and “what will be” in the AI hardware/value stack.
