Qualcomm Says AI Data Centers Will Become Its Next Big Growth Engine

Qualcomm outlined a major AI data-center expansion plan, targeting more than $15 billion in data-center revenue by fiscal 2029 and lifting its non-handset revenue target to $40 billion. The company also announced a planned $3.9 billion acquisition of Modular and named Meta and Microsoft as early hyperscaler customers, with Meta’s CPU deployment slated for 2H 2028. The update reinforces Qualcomm’s diversification beyond smartphones and could materially re-rate the stock if customer adoption accelerates.

Analysis

The market is likely underestimating how much this changes QCOM’s multiple if management can prove even modest design wins in AI infrastructure. The real second-order effect is not just incremental revenue; it is that Qualcomm is trying to re-rate from a handset-cycle proxy to a diversified compute platform, which would compress the valuation gap versus AVGO/MRVL on forward revenue durability. The presence of META and MSFT matters less as near-term dollars and more as third-party validation that QCOM can clear the credibility hurdle with hyperscalers, a customer class that usually de-risks follow-on adoption once one flagship deployment works.

The key bear case is timing mismatch. Even if the product/partner story is real, meaningful revenue may still sit years out, while investor expectations can outrun deployment cadence in the next 2-4 quarters. That creates a classic setup where the stock can rally on narrative but remain vulnerable to any slip in software integration, silicon tape-out, or qualification cycles; in data-center semis, one missed validation window can defer monetization by 6-9 months and compress implied NPV sharply.

The most interesting implication is competitive pressure on the rest of the ecosystem. If QCOM gets traction with power-efficient CPUs and a software portability layer, the marginal loser is not NVDA first; it is the cluster of second-tier custom silicon and networking vendors that depend on being ‘good enough’ rather than best-in-class. The contrarian read is that the market may be too focused on whether QCOM can beat NVDA head-on, when the more probable path is that it wins share in efficiency-sensitive, cost-constrained deployments where ROI is judged on watts, cooling, and total platform cost rather than peak performance.

Key Decisions for Investors

• Buy QCOM on pullbacks over the next 2-6 weeks; use a staged entry because the first upside leg is likely narrative-driven while the second leg depends on customer milestones. Risk/reward is attractive if the market starts pricing even a mid-single-digit revenue probability on the 2029 data-center target.
• Pair trade: long QCOM / short MRVL over 3-12 months. If Qualcomm’s software layer and named hyperscaler traction are real, it should draw differentiation away from more commodity-exposed infrastructure plays; downside is if QCOM fails to convert pilots and the pair mean-reverts.
• Buy a medium-dated QCOM call spread into the next catalyst cycle; structure for a 3-6 month window to express upside from re-rating without paying full delta for a long-duration execution story. Best if implied vol remains moderate after the announcement.
• Fade NVDA only tactically, not structurally. QCOM is more likely to pressure lower-tier custom-silicon winners than the market leader; any short NVDA position should be small and event-driven, with a tight stop, because the core thesis here is share expansion at the edge of the market, not a regime change in accelerator demand.
• Watch META deployment milestones as the cleanest confirmatory catalyst; if qualification or production timing slips, cut QCOM exposure quickly. This is a 6-18 month execution story, not a one-quarter trade.