Alphabet Inc. (NASDAQ:GOOGL) is putting another $15 billion behind the AI infrastructure race.
Reuters reported September 9 that Google plans to invest at least €13 billion, or roughly $15.1 billion, in Finnish AI infrastructure over the next two years. The project includes three new data centers and a 22-year agreement to purchase as much as half the output of one of Finland’s nuclear plants. Alphabet has increased expected 2026 capital expenditures to between $195 billion and $205 billion.
For NVIDIA Corporation (NASDAQ:NVDA), the simplest interpretation is bullish. AI infrastructure demand continues to expand at a pace that would have looked absurd a few years ago. Data centers require accelerators, networking and supporting infrastructure, all markets where Nvidia participates.
Google’s Capex Does Not All Flow to Nvidia
But investors should not assume Google’s entire capex budget becomes Nvidia revenue. Google is also one of the world’s most sophisticated designers of custom AI accelerators. Its TPU strategy gives Alphabet leverage over Nvidia and allows it to route appropriate workloads onto internally designed silicon.
Alphabet Inc. has a bull case built around owning more of the stack. Google controls frontier models, cloud distribution, custom accelerators, data centers and some of the world’s largest consumer products. If Gemini demand expands across Search, YouTube, Workspace and Cloud, infrastructure spending can support multiple revenue streams.
The bear case is capital intensity. Spending approaching $200 billion annually creates an enormous return hurdle. Cheap power and proprietary silicon help, but investors still need evidence that AI revenue and productivity gains justify the depreciation and financing burden.
Nvidia Still Benefits From a Larger Pie
For NVIDIA Corporation, Google illustrates both sides of hyperscaler AI spending. The infrastructure boom expands the addressable market, but the largest customers have the strongest incentives to develop alternatives to Nvidia GPUs. Nvidia must therefore keep improving total system performance fast enough that customers continue buying its products even when they possess credible internal options.
Hedge-fund sentiment strengthened for both companies in Q2. Alphabet ownership rose to 275 funds from 265, while Nvidia climbed to 285 from 275. Short interest remains unusually low, at roughly 1.2% of Alphabet’s Class A float and 1.2% of Nvidia’s float as of August 14.
Alphabet arguably offers the more diversified setup because its AI exposure sits inside profitable advertising, cloud and subscription businesses rather than depending entirely on semiconductor demand. Nvidia remains the cleaner infrastructure beneficiary but commands a premium for that purity. Google’s Finland investment is bullish for the entire AI hardware ecosystem, including Nvidia, but Google’s ownership of models, distribution, infrastructure and custom silicon gives investors several ways to win if AI economics migrate across the stack.
READ NEXT:
The Trillion-Dollar AI Capex Gap: Why Too Much Hardware Could Be Nvidia’s Trap and Microsoft’s Opportunity and Redditors Are Skeptical of SpaceX’s Orbital Data Centers. Are They Right?
Hedge Fund Resource Center