Wells Fargo: AI's Data-Center Trade Is Moving Beyond Nvidia
Hey Tigers 🐯!
The AI investment story has largely centered on GPUs and semiconductor leaders. But as AI models and workloads continue to expand, Wall Street is increasingly looking further down the infrastructure chain — toward the companies providing the power and physical capacity needed to run them.
That is the thesis behind $Wells Fargo(WFC)$'s latest initiation of coverage on $APPLIED DIGITAL CORP(APLD)$.
On September 17, 2026, $Wells Fargo(WFC)$ initiated coverage of $APPLIED DIGITAL CORP(APLD)$ with an Overweight rating — $Wells Fargo(WFC)$'s equivalent of a Buy rating — and a $50 price target, naming the company one of its top ideas in the sector. The bank highlighted $APPLIED DIGITAL CORP(APLD)$'s investment-grade-heavy backlog, power-advantaged footprint and what it sees as a material discount to the company's contracted net asset value.
For Wells Fargo, the appeal is not simply that AI demand is growing.
It is that AI needs somewhere to run.
⚡ AI's Next Bottleneck May Be Power
The first phase of the AI boom focused heavily on computing.
Companies needed increasingly powerful GPUs to train and run AI models, creating enormous demand for accelerators, servers and networking equipment.
But GPUs alone are not enough.
Those chips have to operate inside data centers — facilities that require enormous amounts of electricity, cooling, land and grid connectivity.
As AI infrastructure expands, access to power and ready-to-use data-center capacity can therefore become a critical constraint.
That is where $APPLIED DIGITAL CORP(APLD)$ enters $Wells Fargo(WFC)$'s thesis.
The company develops and operates data-center infrastructure designed for high-performance computing, with its portfolio increasingly focused on large-scale AI and hyperscale workloads.
🏗️ Why Wells Fargo Likes Applied Digital
$Wells Fargo(WFC)$'s thesis is based on several specific characteristics of $APPLIED DIGITAL CORP(APLD)$'s existing portfolio.
1.More than 1.4 GW of contracted capacity
As of the latest reported portfolio data, Applied Digital has secured more than 1.4 gigawatts of contracted capacity across five data-center campuses in three states.
That distinction matters.
This is not simply a collection of proposed facilities waiting for customers. The capacity is tied to contracted leases, giving investors greater visibility into the company's development pipeline.
$APPLIED DIGITAL CORP(APLD)$'s latest disclosures indicate that these leases are structured with long initial terms and take-or-pay provisions, meaning customers remain obligated for contracted capacity even if they do not fully utilize it.
2.Major hyperscale counterparties
More than 70% of $APPLIED DIGITAL CORP(APLD)$'s contracted megawatts are tied directly to investment-grade counterparties $Meta Platforms, Inc.(META)$ and $Oracle(ORCL)$, according to reporting on $Wells Fargo(WFC)$'s initiation.
That is important because large-scale AI infrastructure requires substantial upfront investment.
Long-term commitments from major technology companies can provide greater visibility into future revenue and help support the financing required to build additional capacity.
3.A power-advantaged footprint
$Wells Fargo(WFC)$ also highlighted $APPLIED DIGITAL CORP(APLD)$'s power position and what it considers relatively lower regulatory risk compared with some peers.
For data-center developers, having access to suitable power is not a minor operational detail.
It can be a competitive advantage.
Securing land is one challenge. Building the facility is another. But obtaining enough grid capacity and bringing that power online can be one of the hardest constraints to overcome.
🧠 Why AI Infrastructure Is Becoming an Investment Theme
The AI investment ecosystem is much larger than the companies making AI chips.
Think of the infrastructure chain:
AI models
↓
GPUs & semiconductors
↓
Servers & networking
↓
Data centers
↓
Power & grid infrastructure
$NVIDIA(NVDA)$ sits near the center of the computing layer.
$APPLIED DIGITAL CORP(APLD)$ is targeting the infrastructure underneath it.
That does not mean $APPLIED DIGITAL CORP(APLD)$ is replacing $NVIDIA(NVDA)$ in the AI trade.
Instead, $Wells Fargo(WFC)$'s research highlights another way investors can think about AI's capital cycle: the growth of computing demand creates demand for the physical infrastructure required to support that computing.
💰 The Valuation Argument
$Wells Fargo(WFC)$'s thesis also revolves around $APPLIED DIGITAL CORP(APLD)$'s contracted net asset value.
According to the bank, $APPLIED DIGITAL CORP(APLD)$'s contracted portfolio represents substantial underlying value, while the stock has been trading at a material discount to that contracted NAV. $Wells Fargo(WFC)$ estimates the contracted book alone at roughly $30 per share on a net-present-value basis, creating what it sees as an opportunity for a valuation re-rating.
The logic is important.
$APPLIED DIGITAL CORP(APLD)$ is not being valued purely on the revenue it generates today. Investors are also assigning value to infrastructure that has already been contracted and is expected to come online in the future.
That creates a potential gap between:
Contracted infrastructure value and What the market currently assigns to the stock.
$Wells Fargo(WFC)$'s $50 target therefore reflects not only expectations for AI infrastructure demand, but also the possibility that the market increasingly recognizes the value of $APPLIED DIGITAL CORP(APLD)$'s contracted portfolio.
📊 From GPUs to Gigawatts
The broader takeaway from $Wells Fargo(WFC)$'s initiation is that AI infrastructure can be viewed through a different metric.
Instead of asking only:
How many GPUs are being shipped?
Investors can also ask:
How many megawatts of power and data-center capacity are available to support those GPUs?
For companies like $APPLIED DIGITAL CORP(APLD)$, gigawatts become an important part of the investment story.
The company's portfolio has grown to more than 1.4 GW of contracted capacity across five campuses, while its customers include major technology companies with substantial AI infrastructure needs.
That gives Wells Fargo a way to frame $APPLIED DIGITAL CORP(APLD)$ not simply as a data-center developer, but as a potential infrastructure provider for the next stage of AI expansion.
⚠️ The Real Test: Execution
The AI infrastructure opportunity does not eliminate execution risk.
$APPLIED DIGITAL CORP(APLD)$ still needs to turn contracted capacity into operating data centers, which requires substantial capital expenditure, construction and successful power delivery.
And while long-term contracts with major customers can improve revenue visibility, they do not remove risks related to project execution, financing, customer concentration or the timing of capacity coming online.
This distinction matters.
Contracted capacity is not the same as completed operating capacity.
$Wells Fargo(WFC)$'s thesis ultimately depends on $APPLIED DIGITAL CORP(APLD)$'s ability to convert its existing pipeline into infrastructure, revenue and cash flow.
For investors looking beyond the traditional AI-chip trade, $APPLIED DIGITAL CORP(APLD)$ offers exposure to a different part of the AI infrastructure chain. Whether that translates into the valuation $Wells Fargo(WFC)$ expects will depend on what happens after the contracts are signed: how quickly the capacity gets built, funded and put to work. 🐯
💬 Your Turn: Join the Discussion
$Wells Fargo(WFC)$ just initiated $APPLIED DIGITAL CORP(APLD)$ at Overweight with a $50 target, betting that power and data-center capacity — not just GPUs — will be the next AI bottleneck. What's your read?
A. Real signal — infrastructure/power plays are the next leg of the AI trade
B. Too early — contracted capacity ≠ built capacity, execution risk is still the whole story
C. Somewhere in between — interesting thesis, but wait for projects to actually come online
Share your view below — thoughtful comments may receive Tiger Coins! 🪙
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.
Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.
I like the AI power + data-center infrastructure theme, but I would not treat APLD as a safe investment yet.
Why?
AI needs more data centers and electricity.
APLD has large contracted capacity and big customers.
Long-term contracts can give better revenue visibility.
But contracted capacity ≠ completed data centers.
Construction, financing, power delivery and timing are major risks.
Customer concentration is another risk.
The $50 Wells Fargo target is an analyst view, not a guarantee.
What I would watch:
Contracts → Construction → Power online → Revenue → Cash flow
If APLD successfully converts its contracts into operating data centers and strong cash flow, the story becomes much stronger.
Bottom line: C. Interesting AI infrastructure play, but I would wait for execution proof rather than chase the stock.
APLD has about 1.4 GW contracted across five campuses, representing roughly $36B of contracted revenue over initial 15-year terms. But only 175 MW at Polaris Forge 1 was live as of May 31, 2026.
That gap is the real investment question. The backlog provides visibility, while the huge capex and roughly $5B debt also show why execution and financing matter. FY2026 adjusted EBITDA reached $107M, but the company still reported a $249M GAAP net loss.
So I’d watch MW delivered, construction timelines, funding costs and free cash flow more closely than the $50 target. If contracted capacity keeps converting into productive assets, the thesis gets stronger; if construction slips, the NAV discount may be justified.
For me, APLD is less a GPU story and more a “can management turn gigawatts into durable cash flow?” story.
@Capital_Insigh