$4.8 Trillion and Climbing

Alphabet just crossed $4.8 trillion in market cap. Nvidia sits at $5.2 trillion. The gap is closing fast.

A year ago, this race wasn’t even a conversation. Google was the “search company with an AI problem.” Now it’s the integrated AI powerhouse that Wall Street can’t stop buying. The stock is up 160% in twelve months.

What changed? Everything — and nothing. Google always had the pieces. It just took the market a while to notice they’d assembled them.

The Full Stack Advantage

Here’s what Alphabet owns that nobody else does: the entire AI stack, top to bottom.

Silicon. Google’s TPUs aren’t a side project anymore. TPU revenue exploded from $3 billion to $25 billion in two years. That’s not a rounding error — that’s a business that would be a top-tier semiconductor company on its own.

Models. Gemini isn’t just competitive with GPT-5 and Claude — it’s deeply integrated into every Google product touching 2 billion users. That distribution advantage is almost impossible to replicate.

Cloud. Google Cloud crossed the profitability inflection point and keeps accelerating. Enterprise AI workloads are flooding in.

Distribution. Search, YouTube, Android, Workspace. Two billion users who wake up inside Google’s ecosystem every morning.

No other company has all four layers. Nvidia makes chips but depends on others for everything above. Microsoft has cloud and distribution but rents its models from OpenAI. Meta has distribution and models but no cloud business. Apple has distribution but barely has AI.

Alphabet has everything.

The Anthropic Play Nobody Understands

Google’s relationship with Anthropic is the most misunderstood deal in tech.

The numbers: Google has committed over $200 billion in cloud credits and investment to Anthropic. In return, Anthropic runs on Google Cloud. That deal now represents roughly 40% of Google Cloud’s backlog.

Critics call it circular. Money goes out, comes back as cloud revenue. And they’re not wrong — it is circular. But that misses the point.

Google isn’t buying Anthropic’s friendship. It’s buying workload lock-in. Every Anthropic model trained on TPUs is a model that can’t easily move to AWS. Every enterprise customer accessing Claude through Google Cloud is a customer that stays on Google Cloud.

The circularity is the feature, not the bug.

TPUs: The Quiet Revolution

The TPU story deserves its own spotlight. Three billion to twenty-five billion in revenue. That’s an 8x jump.

What’s driving it: hyperscalers and AI labs discovered that for training large models, TPUs offer better price-performance than Nvidia’s GPUs for many workloads. Google kept this advantage in-house for years. Now it’s selling capacity to anyone willing to pay.

The irony is thick. Nvidia’s dominance created the opening. GPU scarcity pushed companies to explore alternatives. Google’s TPUs were sitting right there, battle-tested on the most demanding AI workloads in the world.

Now TPU revenue is growing faster than Nvidia’s data center business in percentage terms. From a lower base, sure. But the trajectory is unmistakable.

Why Integrated Players Win

The AI era rewards vertical integration. This shouldn’t be surprising — it’s the same pattern that made Apple the world’s most valuable company in the mobile era.

When you control the silicon, the models, the cloud, and the distribution, you capture margin at every layer. You move faster because you don’t negotiate across company boundaries. You optimize end-to-end in ways that modular players can’t.

Nvidia makes incredible chips. But it captures value at one layer. Alphabet captures value at every layer.

That’s why the market cap gap is closing.

What $5 Trillion Looks Like

Alphabet doesn’t need a miracle to pass Nvidia. It needs continued execution on what’s already working: TPU growth, Cloud growth, Gemini integration, and the Anthropic partnership delivering cloud backlog.

The bear case requires Google to fumble execution across multiple fronts simultaneously. Possible, but not the base case when every metric is accelerating.

The world’s largest company might not be a chipmaker for much longer. It might be the company that owns the whole stack.

The market is starting to price that in. The question is whether it’s pricing it in fast enough.