The AI trade is much bigger than $NVIDIA(NVDA)$ .
If you look at the entire AI infrastructure stack, the money is flowing through multiple layers — from compute and cloud capacity all the way to networking, power and data centers.
Here’s how I’m mapping it:
☁️ AI Clouds
$SpaceX(SPCX)$ $NEBIUS(NBIS)$ $CoreWeave, Inc.(CRWV)$ $IREN Ltd(IREN)$ $SharonAI Holdings Inc.(SHAZ)$
This is the layer closest to the actual demand for AI compute.
As AI models become larger and inference workloads grow, companies need more GPU capacity and more specialized infrastructure. The AI cloud names are essentially selling access to that compute.
🏢 Hyperscalers
$Microsoft(MSFT)$ $Meta Platforms, Inc.(META)$ $Alphabet(GOOGL)$ $Amazon.com(AMZN)$ $Oracle(ORCL)$
These companies sit at the other end of the spectrum.
They have massive existing cloud infrastructure, huge customer bases and the capital to keep spending on AI data centers.
The key question here isn't simply who has the most GPUs.
It’s who can turn enormous AI capex into actual revenue, cloud demand and eventually cash flow.
💻 Enterprise AI
$Palantir Technologies Inc.(PLTR)$ $ServiceNow(NOW)$ $Salesforce.com(CRM)$ $Snowflake(SNOW)$ $MongoDB Inc.(MDB)$ $CrowdStrike Holdings, Inc.(CRWD)$ $Palo Alto Networks(PANW)$ $Cloudflare, Inc.(NET)$ $DigitalOcean Holdings, Inc.(DOCN)$ $Innodata(INOD)$
This is where AI infrastructure starts turning into software.
The winners here don't necessarily need to build the underlying compute.
They need businesses to spend more money using AI — for data, security, automation, development and enterprise workflows.
That creates a different way to participate in the AI cycle.
🧠 Semiconductors
$NVIDIA(NVDA)$ $Micron Technology(MU)$ $SK hynix(SKHY)$ $Broadcom(AVGO)$ $Taiwan Semiconductor Manufacturing(TSM)$ $Intel(INTC)$ $ARM Holdings(ARM)$ $Advanced Micro Devices(AMD)$ $Marvell Technology(MRVL)$ $Samsung Electronics Co., Ltd.(SSNLF)$
This is still the core of the stack.
GPUs get most of the attention, but AI systems also need memory, networking silicon, custom accelerators and advanced manufacturing.
That’s why the semiconductor opportunity is much broader than one company.
🌐 Networking
$Arista Networks(ANET)$ $Astera Labs, Inc.(ALAB)$ $Coherent(COHR)$ $Lumentum(LITE)$
As clusters get larger, moving data between GPUs becomes increasingly important.
More compute means more bandwidth.
More bandwidth means more optical connectivity, switching and high-speed interconnects.
This is one of the less obvious ways to play the continued expansion of AI clusters.
🖥️ Servers
$Dell Technologies Inc.(DELL)$ $Hewlett Packard Enterprise(HPE)$ $SUPER MICRO COMPUTER INC(SMCI)$
Someone still has to put all those chips into actual systems.
AI servers combine GPUs, CPUs, memory, networking and cooling into deployable infrastructure.
The growth here is tied directly to how quickly hyperscalers and AI clouds are building capacity.
⚡ Power
$Bloom Energy Corp(BE)$ $Vertiv Holdings LLC(VRT)$ $Eaton Corp PLC(ETN)$ $GE Vernova Inc.(GEV)$ $Quanta(PWR)$
This is where the AI infrastructure story gets really interesting.
A data center doesn't run on GPUs alone.
It needs electricity, backup systems, power distribution and cooling.
If AI demand keeps driving larger data centers, power infrastructure becomes a constraint — and potentially a major investment theme of its own.
🏭 Data Centers
$Digital Realty Trust Inc(DLR)$ $Core Scientific, Inc.(CORZ)$ $APPLIED DIGITAL CORP(APLD)$ $TeraWulf Inc.(WULF)$ $Cipher Mining Inc.(CIFR)$ $Hut 8 Mining Corp(HUT)$ $CleanSpark, Inc.(CLSK)$ $Riot Platforms(RIOT)$
And then there is the physical real estate.
AI needs somewhere to live.
That means land, buildings, grid connections, cooling and enormous amounts of power.
Some of these companies started with Bitcoin mining and are now positioning their infrastructure for AI/HPC workloads.
The bigger picture looks something like this:
AI demand → compute → chips → servers → networking → power → data centers
And above that infrastructure sits the software layer that actually monetizes the compute.
That’s why I don't think the AI trade should be viewed as a single-stock story.
The more AI gets deployed, the more spending gets pushed through the entire ecosystem.
The question is simply which layer captures the economics at each stage of the cycle. 👀
Save this map. There are a lot more ways to play AI than just buying the obvious names.
Markets are always moving - and sometimes, the best move is knowing what works for you.
With Treasury yields, oil prices and rate expectations keeping markets on edge this week, investors are once again thinking carefully about where to position next. There’s no one-size-fits-all choice in investing — and the same goes for Tiger Merch. This month’s hot picks are in, featuring the Tiger Toiletry Bag, Universal Travel Adapter, Tiger Umbrella and more favourites chosen by fellow Tigers.
Explore the Monthly Hot Picks in Tiger Coin Mall, now 12% OFF for a limited time.
Comments