#AIStockGuruReportedlyBullishOnAI

AI IS ENTERING ITS INFRASTRUCTURE ERA — AND THE NEXT WINNERS MAY BE HIDING BEHIND THE HEADLINES

The AI investment story is changing.

For the past few years, the market has been obsessed with one simple question:

“Who is building the most powerful AI?”

Now a more important question may be emerging:

“Who is supplying everything AI needs to keep growing?”

That distinction matters.

A reported rebuilding of AI positions after a painful 67% drawdown has attracted attention, especially because the names reportedly being watched are not concentrated in only one part of the AI ecosystem.

The basket includes SNDK, BE, INTC, CRWV, SKHY and AMD.

At first glance, these companies may look unrelated.

Look deeper, however, and a common theme starts to appear:

AI infrastructure.

THE 67% DRAWdown CHANGES THE STORY

A 67% drawdown is significant.

It can completely change how an investor approaches risk, position sizing and capital allocation.

After a major decline, simply rebuilding positions is not necessarily a signal that the previous strategy is being repeated.

The more important signal may be the selection of new positions.

Where is capital moving?

Which parts of the AI ecosystem are attracting attention?

And what does that positioning suggest about the next stage of the AI cycle?

That is where this reported basket becomes interesting.

THE AI SUPPLY CHAIN IS MUCH BIGGER THAN GPUs

When people think about AI, GPUs usually come first.

That makes sense because powerful accelerators are essential for training and running advanced AI models.

But GPUs cannot operate in isolation.

Every major AI data center also needs:

High-performance memory.

Massive storage capacity.

Reliable electricity.

Advanced cooling.

Networking infrastructure.

Cloud computing.

Data-center capacity.

Semiconductor manufacturing.

Specialized processors.

This means AI expansion creates demand across an entire industrial ecosystem.

The biggest opportunity may not always be the company everyone is already talking about.

Sometimes the more interesting opportunity sits one or two layers deeper in the supply chain.

SNDK: THE STORAGE CONNECTION

Storage is one of the less glamorous parts of the AI narrative, but it is extremely important.

AI systems generate and process enormous quantities of data.

Training datasets, model outputs, applications and enterprise workloads all require storage infrastructure.

That makes storage companies relevant to the broader AI build-out.

SNDK therefore represents an interesting second-layer AI exposure.

The thesis is not simply:

“AI goes up, therefore storage goes up.”

It is about understanding the infrastructure requirements created by expanding AI workloads.

More computing generally means more data.

More data means greater storage requirements.

That connection deserves attention.

SKHY AND THE MEMORY BOTTLENECK

Memory could be one of the most important pieces of the puzzle.

Advanced AI accelerators require extremely fast access to data.

High-bandwidth memory helps move large quantities of information efficiently between processors and memory.

As AI models become larger and computing workloads become more demanding, memory becomes increasingly strategic.

That makes SK Hynix an important name to watch within the AI supply chain.

The broader lesson is simple:

A powerful processor is only as useful as the infrastructure supporting it.

If memory becomes a bottleneck, the entire AI expansion cycle can face limitations.

POWER MAY BECOME THE NEXT BIG AI TRADE

There is another problem that cannot be solved with software alone.

Electricity.

AI data centers require enormous amounts of energy.

As companies build larger clusters and deploy more powerful computing systems, electricity demand becomes a critical infrastructure question.

This brings BE into the conversation.

Bloom Energy represents the power side of the AI infrastructure equation.

The long-term AI story therefore extends beyond chips.

It also includes energy generation, reliability and infrastructure capable of supporting the next generation of data centers.

INTEL: A DIFFERENT KIND OF AI BET

Intel brings a different angle.

Its importance is connected not only to AI demand but also to semiconductor manufacturing and foundry ambitions.

If Intel can improve its manufacturing capabilities and successfully attract external customers, it could become an important participant in the semiconductor supply chain.

That makes INTC a potential turnaround component within the broader AI infrastructure thesis.

But turnaround stories require patience.

Execution matters.

Technology matters.

Capital requirements matter.

And competition remains intense.

AMD: THE COMPETITION LAYER

AMD represents another major part of the equation.

The AI accelerator market is not necessarily going to remain a one-company story forever.

Competition matters because customers want multiple suppliers, better performance and potentially more attractive economics.

AMD therefore provides exposure to the competitive side of AI computing.

Investors are watching product roadmaps, market share, margins, supply availability and customer demand.

The AI market can grow rapidly while competition simultaneously becomes more aggressive.

That creates both opportunity and risk.

CRWV: COMPUTE AS A SERVICE

Then there is CoreWeave.

AI models require massive computing capacity, but not every company wants to build and operate its own infrastructure.

Specialized cloud providers can help supply that capacity.

This creates another important link in the chain:

AI applications create demand.

Cloud infrastructure provides access to computing.

Accelerators perform the heavy workloads.

Memory feeds the processors.

Storage manages enormous quantities of data.

Power keeps everything running.

Manufacturers produce the underlying hardware.

Suddenly, the AI ecosystem looks much larger than a simple GPU trade.

THE SECOND-DERIVATIVE AI OPPORTUNITY

This may be the most important takeaway.

The first AI investment wave focused heavily on direct beneficiaries.

The second wave could increasingly focus on the companies that enable those beneficiaries to scale.

Think about it like an ecosystem.

AI models need compute.

Compute needs chips.

Chips need memory.

AI workloads need storage.

Data centers need power.

Cloud companies need infrastructure.

Semiconductor companies need manufacturing capacity.

Every layer creates another potential investment narrative.

That is what makes the reported basket so interesting.

GLOBAL MARKETS ADD ANOTHER DIMENSION

There is also a geographical component.

SK Hynix is deeply connected to the Asian semiconductor ecosystem, while companies such as Intel, AMD, Bloom Energy and CoreWeave are associated with U.S. markets.

This creates a global AI infrastructure cycle.

News can emerge in Asia before U.S. markets open.

Semiconductor developments can move memory stocks first.

Later, U.S. markets may react through chip, cloud, energy and infrastructure companies.

For crypto-native traders, Gate's Stock Perps can provide access to selected stock-market exposure using USDT, creating another way to follow these cross-market narratives.

The opportunity is not about eliminating risk.

It is about understanding how capital can rotate across different markets and sessions.

WHAT I WOULD WATCH NEXT

I would not focus only on headlines.

I would watch:

Price structure.

Trading volume.

Relative strength.

Memory pricing.

AI capital expenditure.

Hyperscaler spending.

Data-center construction.

Energy demand.

Semiconductor supply.

Earnings expectations.

And the rotation between major AI leaders and second-layer infrastructure names.

If money continues moving deeper into the AI supply chain, the infrastructure thesis could become increasingly important.

THE BIGGER PICTURE

The reported rebuilding of AI positions after a 67% drawdown is interesting.

But the real story may not be the recovery itself.

It may be the composition of the basket.

SNDK represents storage.

SKHY represents advanced memory.

BE represents power.

INTC represents manufacturing and turnaround potential.

AMD represents accelerator competition.

CRWV represents specialized cloud infrastructure.

Together, they form a broader picture of what AI needs to scale.

This is why I believe the next phase of the AI market could become less about simply finding the “next AI winner” and more about understanding the infrastructure behind the entire industry.

The AI revolution needs intelligence.

But intelligence needs hardware.

Hardware needs memory.

Memory needs manufacturing.

Computing needs storage.

Data centers need power.

And all of it needs infrastructure.

That could be the real second-layer AI trade.

The biggest opportunity may not always be the company building the AI.

Sometimes, it may be the company building everything that allows AI to exist at massive scale.

#weeklyshare @Gate_Square #ShareWeekly #GateSquare