AI bubble or AI revolution? NVIDIA’s earnings report provides the answer

It’s that it once again proves:

The AI infrastructure revolution is still in its early stage.

In the past, the market asked:

“Is AI a bubble?”

Now the market is starting to ask:

“If AI isn’t a bubble, then how big will the compute demand be in the next ten years?”

The answer could be far beyond what most people imagine.

1. NVIDIA is becoming the “infrastructure company” of the AI era 💰

Many people still see NVIDIA as a chip company.

But today’s NVIDIA is no longer just selling GPUs.

It sells:

• AI computing platforms
• data center solutions
• the CUDA ecosystem
• network interconnection
• software services

GPUs are just the entry point.

The real value is the entire AI infrastructure ecosystem.

In the past 20 years:

The internet expanded through server growth.

In the mobile internet era, expansion came through smartphones.

Over the next 10 years:

AI will expand through the growth of compute infrastructure.

And NVIDIA is occupying the most core position in this era.

2. The biggest change: AI demand hasn’t ended—it’s entering a new phase 🔥

In the past two years:

AI demand mainly came from large-model training.

Microsoft, Google, Meta, and Amazon were feverishly buying GPUs.

The market worried:

“After large-model training ends, what happens to GPU demand?”

But the market ignored one point:

Training is only the beginning.

The real big market is inference.

In the future:

Every AI Agent,

every robot,

every autonomous driving system,

every enterprise AI assistant,

will need to continuously consume compute power.

In the future, compute demand will no longer come only from humans.

It will come from billions of AI agents.

That is NVIDIA’s true long-term logic.

3. The AI industry chain is entering a golden cycle 🌊

NVIDIA’s rise isn’t just NVIDIA’s story.

Behind it is an entire industry chain:

1. GPU

NVIDIA continues to hold a core position.

AMD is competing.

But the ecosystem advantage remains huge.

2. HBM memory

The stronger the AI chip, the higher the demand for high-speed memory.

In the future, the bottleneck of AI might not be GPUs.

It could be:

HBM production capacity.

Benefiting directions:

• SK hynix
• Samsung
• Micron

3. Advanced packaging

Future competition won’t be about a single chip.

It will be:

GPU + HBM + packaging + networking.

Advanced packaging is becoming a new strategic resource.

4. Power and data centers

This is the biggest direction that the market is underestimating.

AI needs compute.

Compute needs power.

The future AI revolution is, at its core, also an energy revolution.

4. But investors must pay attention: even the best company may come with the highest price ⚠️

NVIDIA’s fundamentals remain strong.

But the stock market trades expectations.

What the market will focus on next:

First:

How long can AI capital expenditures continue?

If Microsoft, Google, and Meta keep expanding investment, the AI cycle continues.

If capex slows down, the whole sector may see its valuation reset.

Second:

When will AI commercialization be realized?

If AI truly improves enterprise efficiency:

This investment cycle could last for more than ten years.

If it’s only model competition:

the market may price in the future too early.

Third:

Will competitive dynamics change?

NVIDIA’s biggest moat:

isn’t the chips.

It’s the ecosystem.

But in the coming years:

competition from custom AI chips, ASICs, AMD, and others will add pressure.

5. The real big opportunity isn’t predicting NVIDIA’s next day’s up or down 📈

It’s finding out:

who will become the infrastructure winner of the AI era.

Every tech revolution has similar patterns:

Railroad era:

the winners were railroads and steel.

Internet era:

the winners were cloud computing and platforms.

AI era:

the winners will be:

compute,

energy,

storage,

networking,

robots.

Conclusion 🌱

The biggest signal released by NVIDIA’s earnings report:

The AI revolution hasn’t ended—it’s shifting from “model competition” into the “infrastructure building era.”

In the short term:

valuations will fluctuate.

capital will trade expectations.

the stock price may adjust.

But in the long term:

AI is becoming a new foundation for productivity.

The biggest industrial opportunity over the next ten years

may not be who creates a bigger model.

It will be who controls the infrastructure that keeps the entire AI world running.

In the gold-rush era, the truly great companies are often not the ones who dig for gold mines, but the ones who provide the tools.

Today, NVIDIA

is becoming the most important “shovel seller” of the AI era. 🚀

#Summer Creative Camp