By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
While Qualcomm's pivot to data centers and the Meta deal offer potential, the panelists agree that execution risks, particularly ramping up production and displacing Nvidia, are significant. The success of this shift hinges on winning multiple customers and maintaining favorable margins.
Risk: Ramping up production and displacing Nvidia's dominance in AI compute and software ecosystems.
Opportunity: Securing multiple cloud customers and improving data-center margins to shift EPS growth profile.
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
Qualcomm (NASDAQ: QCOM) has spent years trying to convince investors it can be more than a smartphone-chip company. At its investor day on Wednesday, it made its boldest case yet. The company nearly doubled its target for non-handset revenue in fiscal 2029, raising the goal to about $40 billion from $22 billion. And for the first time, it put hard numbers behind its data center ambitions, calling for more than $15 billion in data center revenue by that same year.
Investors liked what they heard, and shares jumped sharply on the news, rising as much as 15%.
Where to invest $1,000 right now? Our analyst team just revealed what they believe are the 10 best stocks to buy right now, when you join Stock Advisor. See the stocks »
The figures are bold for a company whose chips still sit mostly inside phones. But raising a target is the easy part. The harder question is whether Qualcomm, a relative latecomer to the data center, can build a business of this size in a market that Nvidia already dominates.
Qualcomm's diversification push isn't new, but the scale of it is. The company's prior $22 billion non-handset goal, set in 2024, was already meant to loosen its dependence on smartphones -- a maturing market where it also faces the gradual loss of Apple as a modem customer as the iPhone maker shifts to its own in-house modem products. The new $40 billion target nearly doubles that ambition.
The data center, of course, is the centerpiece of the company's growing ambition. Qualcomm detailed a server processor called the Dragonfly C1000, built around more than 250 of its custom cores, along with a line of artificial intelligence (AI) accelerators designed to run AI models rather than train them. Management is targeting more than $15 billion in data center revenue by fiscal 2029 -- up from almost nothing today.
The most important validation came from a customer. Meta Platforms agreed to a multi-year, multi-generation deal to use Qualcomm's new processor in its data centers, with production starting in the second half of 2028. For a company trying to prove it belongs in the data center, landing one of the world's biggest spenders on computing infrastructure is a meaningful endorsement.
Qualcomm's other growth bets are further along. Its automotive revenue rose 38% year over year to a record $1.3 billion in its fiscal second quarter of 2026 (the period ended March 29, 2026), and management is targeting $10 billion in annual automotive revenue by fiscal 2029, backed by a design-win pipeline it now pegs at about $65 billion.
Still, the targets are a bet, not a result.
Qualcomm is arriving late to a data center market where Nvidia controls the vast majority of AI chip sales and where a deep software ecosystem keeps customers from switching. Qualcomm's HBC-based AI250 accelerator won't begin commercial sampling until mid-2027, and the Meta CPU production doesn't begin until the second half of 2028.
A lot can change between 2026 and the end of 2028.
The current numbers are a reminder of how far the company has to go. Qualcomm's fiscal Q2 revenue was $10.6 billion, and handset chips still accounted for the largest piece at about $6 billion. Data center revenue is a rounding error by comparison. The $40 billion goal assumes years of strong execution in markets where Qualcomm hasn't yet proven it can win at scale.
What makes the stock interesting, however, is that investors aren't paying much for any of this. Qualcomm's reported fiscal Q2 earnings were inflated by a one-time tax benefit, but on a non-GAAP (adjusted) basis the stock trades at about 17 times earnings -- well below the broader market and a fraction of what pricier AI chip names command.
The market, in other words, is treating Qualcomm as a mature smartphone-chip supplier and assigning little value to the data center business it just sketched out.
That mix of a modest valuation and a credible, if unproven, growth story is what makes Qualcomm worth a closer look. Sure, I wouldn't buy the stock on the strength of a 2029 target alone, and the competitive risks in the data center are real. But the Meta agreement suggests the ambition is more than a slide in an investor presentation -- and at this valuation, investors aren't being asked to pay up for a diversification story that finally seems to be taking shape.
Before you buy stock in Qualcomm, consider this:
The Motley Fool Stock Advisor analyst team just identified what they believe are the 10 best stocks for investors to buy now… and Qualcomm wasn’t one of them. The 10 stocks that made the cut could produce monster returns in the coming years.
Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, you’d have $382,359! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,201,390!
Now, it’s worth noting Stock Advisor’s total average return is 883% — a market-crushing outperformance compared to 205% for the S&P 500. Don't miss the latest top 10 list, available with Stock Advisor, and join an investing community built by individual investors for individual investors.
**Stock Advisor returns as of June 26, 2026. *
Daniel Sparks and his clients have positions in Apple. The Motley Fool has positions in and recommends Apple, Meta Platforms, Nvidia, and Qualcomm. The Motley Fool has a disclosure policy.
The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.
Four leading AI models discuss this article
"Qualcomm could become a meaningful data-center player, but the path to $15B+ in data-center revenue by 2029 is highly uncertain and depends on winning in a dominated market."
Qualcomm’s investor-day push signals a bold shift toward non-handsets, with a $40B non-handset target and $15B+ data-center revenue by 2029. The Meta deal offers real validation, but the path is long and uncertain: Nvidia currently dominates AI compute and software ecosystems, while Qualcomm’s Dragonfly C1000 won’t sample until mid-2027 and production won’t scale until 2028–29. The ramp requires customers to switch costs and risk, plus sustained capex and favorable margins in a highly competitive, capital-intensive market. Valuation remains modest, but execution risk and the need to displace entrenched competitors temper the upside.
The strongest counter is that Nvidia and peers already own the data-center AI moat; Qualcomm is a late entrant with unproven scale in servers, so the $15B data-center target by 2029 may be wishful thinking unless multiple, large, durable wins materialize around 2028–29. A slower AI cycle or Meta revising its plan could render the target unattainable.
"Qualcomm is currently priced as a stagnant handset supplier, meaning any meaningful revenue realization from their data center pipeline provides a free, high-convexity call option for shareholders."
Qualcomm’s pivot to a $15 billion data center target by 2029 is a classic 'show me' story. While the Meta partnership provides a vital beachhead, we must distinguish between design wins and actual margin-accretive revenue. At a 17x forward P/E, the market is pricing this as a legacy handset play, effectively giving the AI data center narrative a zero-dollar valuation. If they execute, the multiple expansion potential is massive. However, the 2028 production timeline for Meta is an eternity in AI hardware cycles. Qualcomm is essentially betting that inference-specific silicon will become a commodity where their power-efficiency advantage outweighs Nvidia’s software-moat dominance.
The 2028 production timeline is dangerously late; by then, the industry may have shifted from general-purpose inference chips to highly specialized ASICs or optical computing, rendering Qualcomm's 'Dragonfly' architecture obsolete before it hits volume.
"Meta's commitment is credible but insufficient; QCOM needs to prove it can scale to $15B data center revenue with acceptable margins against entrenched competition, not just land one flagship customer."
The Meta deal is real validation, but the article conflates announcement with execution. QCOM trades at 17x forward earnings partly because the market *already* prices in some diversification upside—the stock is up 40% YTD. The $40B non-handset target requires $15B data center revenue by FY2029, but commercial sampling of the AI250 doesn't start until mid-2027, production until H2 2028. That's 18 months to ramp from zero to material scale in a market where Nvidia has 80%+ share and entrenched software moats (CUDA). The Meta win is one customer; we need visibility on pipeline breadth and gross margins before treating this as a done deal.
If QCOM's valuation already reflects modest diversification expectations and the stock has already rallied sharply on the announcement, the risk/reward may be worse than the article implies—you're buying the execution story at a price that's already moved.
"Qualcomm's 2029 targets assume flawless execution in a market where product availability and ecosystem lock-in remain unproven for years."
Qualcomm's jump to a $40B non-handset target by FY2029, including $15B+ from data centers, rests on a Meta multi-year CPU deal starting H2 2028 and Dragonfly C1000 sampling in 2027. Yet current data center revenue is negligible against $10.6B Q2 total, with handsets still dominant at ~$6B. Nvidia's software moat and the 2-3 year lag before meaningful ramps create execution risk that the 17x non-GAAP multiple may not fully discount if automotive and IoT growth also slip.
The Meta commitment and $65B automotive design-win pipeline could accelerate adoption faster than modeled, turning the low multiple into a re-rating catalyst if even partial targets are hit by 2028.
"Dragonfly's ramp risk means a Meta deal alone won't unlock a durable 2029 data-center target without 2–3 large cloud wins and meaningful margin expansion."
Gemini, the 'show me' thesis hinges on a broad, margin-friendly data-center footprint. The real risk you understate is the ramp: Dragonfly sampling in 2027 and production 2028–29 presumes rapid multi-customer adoption and material gross margins, which Nvidia currently dominates on software. A single Meta deal may prove insufficient to re-rate Qualcomm unless 2–3 large cloud customers commit and improve data-center margins meaningfully; otherwise the target stays aspirational.
"The market is mispricing Qualcomm by treating it as a stagnant handset company rather than a latent data center challenger."
Claude, you’re missing the forest for the trees on valuation. The 17x forward P/E isn't pricing in diversification; it's pricing in terminal decline for the handset business. If Qualcomm manages even 30% of that $15B data center target, the EPS growth profile shifts from low-single digits to mid-teens. The risk isn't just execution; it's the 'handset trap'—investors are so anchored to QCOM as a smartphone proxy that they’ll ignore the data center pivot until revenue is actually on the tape.
"Meta's commitment validates Qualcomm's technical roadmap, not necessarily its ability to achieve margin-accretive scale against Nvidia's entrenched moat."
Gemini's 'handset trap' framing is sharp, but conflates two separate problems. Even if investors repriced QCOM away from terminal decline, the 17x multiple assumes the $15B data-center target is achievable. That's not a valuation problem—that's an execution problem. Hitting 30% of target ($4.5B) by 2029 still requires displacing Nvidia in inference at scale. The real question: does the Meta deal prove Qualcomm can win on power efficiency alone, or does it prove Meta is willing to pay a premium for vendor diversification? Those have opposite implications for margin accretion.
"Meta's diversification motive risks forcing Qualcomm into margin-dilutive deals that prevent the data-center target from driving re-rating."
Claude rightly flags that the Meta deal may reflect diversification rather than pure efficiency wins, but the deeper risk is how that motive shapes pricing. If Qualcomm must discount aggressively to secure share against Nvidia, even hitting $4.5B in revenue by 2029 could compress data-center gross margins below the levels needed for EPS re-acceleration. That outcome would leave the 17x multiple anchored to handset decline rather than unlocking the mid-teens growth Gemini expects.
While Qualcomm's pivot to data centers and the Meta deal offer potential, the panelists agree that execution risks, particularly ramping up production and displacing Nvidia, are significant. The success of this shift hinges on winning multiple customers and maintaining favorable margins.
Securing multiple cloud customers and improving data-center margins to shift EPS growth profile.
Ramping up production and displacing Nvidia's dominance in AI compute and software ecosystems.