Qualcomm is pivoting to position itself as an AI data-center CPU vendor, targeting an estimated $200 billion market opportunity.

Qualcomm spent years as the company inside your phone. Now it wants to be the company inside the data center, too.

At its Investor Day on June 24, 2026, the chipmaker laid out a plan to push hard into AI infrastructure, targeting more than $15 billion in data center revenue by fiscal 2029. The pitch resonated: shares jumped more than 12% in after-hours trading. Qualcomm's ambition rests on a projected market of approximately $200 billion for CPUs by fiscal 2029 or 2030.

The centerpiece is a new product family called Dragonfly, anchored by the Dragonfly C1000, a server CPU built for agentic AI workloads. The C1000 uses a multi-chiplet design, stitching together several smaller silicon pieces instead of one monolithic slab. This approach improves manufacturing yields and design flexibility. The specs are aggressive: the chip packs more than 250 Oryon cores and runs at frequencies above 5 GHz, delivering more than double the performance per watt of competing server CPUs.

Performance per watt matters more than raw speed alone. Data centers are increasingly power-limited rather than space-limited. A chip that doubles output on the same power budget effectively unlocks new capacity.

Qualcomm has already secured Meta Platforms as a marquee customer, with a multi-generation agreement to deploy the Dragonfly C1000 in its data center servers beginning in the second half of 2028. Microsoft is also expected to deploy Qualcomm's High Bandwidth Compute chips (HBC), and additional hyperscalers are likely to contribute to Qualcomm's custom-chip revenue before the year ends.

Qualcomm projects $5 billion in data center revenue for fiscal 2027 as a stepping stone to its $15 billion-plus goal. The company also raised its non-handset revenue target for fiscal 2029 to $40 billion, nearly doubling the previous $22 billion guidance.

By fiscal 2029, smartphones will account for only about one-third of Qualcomm's revenue, with the remainder coming from data centers, automotive, IoT, industrial applications, robotics, and edge AI platforms. The server CPU market has long been dominated by Intel and AMD, with Nvidia competing through its Vera chip. Efficient Arm-based designs have been gaining ground as cloud providers seek alternatives to traditional x86 processors. Qualcomm's argument centers on energy efficiency and cost, betting it can win on total economics in power-constrained environments.

The Meta deal carries significant weight. Chip companies frequently announce server ambitions; far fewer arrive with a hyperscaler committed to a multi-generation contract. Combined with Microsoft's expected HBC deployment, Qualcomm has two of the largest compute buyers backing its strategy. The real test comes when the C1000 hits volume production in the second half of 2028. The fiscal 2027 checkpoint of $5 billion in data center revenue will show whether the $15 billion-plus fiscal 2029 target is achievable.