Qualcomm targets $15B in AI data center chips by 2029

Qualcomm sets sights on Nvidia's AI chip stronghold

Qualcomm has laid out an aggressive five-year strategy to capture a significant share of the AI data center chip market, projecting annual sales exceeding $15 billion from AI components by fiscal 2029. CEO Cristiano Amon presented the plan to investors this week, positioning the company as a credible challenger to Nvidia's overwhelming dominance in the sector.

The announcement at Qualcomm's Manhattan investor day included the unveiling of new AI accelerators and central processing units designed specifically for data center workloads. Investors initially responded enthusiastically, sending shares up as much as 15% before broader market volatility tempered those gains by week's end.

Why it matters

Qualcomm's push into AI data centers represents a fundamental shift for a company historically dependent on smartphone chips. Success would diversify revenue streams and position Qualcomm as a power-efficient alternative at a time when data center energy consumption faces increasing scrutiny. The move also intensifies competition in a market where Nvidia currently holds a near-monopoly, potentially giving enterprise customers more options and leverage.

Beyond handsets: A $40 billion diversification bet

The AI data center initiative is part of a broader transformation under Amon, who became CEO in 2021. Qualcomm now projects $40 billion in annual revenue from non-handset businesses by 2029—double the forecast from two years ago. The company has already expanded into automotive technology, including driver-assistance systems and connected-vehicle platforms, as well as chips for smart home devices and wearables.

Amon emphasized that Qualcomm's power-efficient CPU designs could address growing concerns about AI infrastructure's energy demands. "There is pushback, but that's good because it is going to drive the industry to look for alternatives," he said during the presentation.

The Modular acquisition: Taking on CUDA

Central to Qualcomm's strategy is this week's announced $3.9 billion acquisition of Modular, an AI software company. The deal gives Qualcomm a software platform designed to compete directly with Nvidia's CUDA, the development environment that has kept many AI programmers locked into Nvidia's ecosystem. CUDA allows developers to write AI programs that fully leverage Nvidia's graphics processing units, creating a powerful moat around Nvidia's hardware business.

The investor day featured video endorsements from Meta CEO Mark Zuckerberg and Microsoft CEO Satya Nadella, along with executives from Amazon and Google, signaling industry support for Qualcomm's expanded ambitions.

Can Qualcomm execute on multiple fronts?

Amon, a Brazilian engineer who joined Qualcomm three decades ago, acknowledged the scale of the challenge but dismissed concerns about overextension. The company is simultaneously pursuing data center AI chips, deeper automotive market penetration, and competition with Intel and AMD in PC processors.

"We have a very, very strong engineering culture. It's a company that is not afraid of taking on completely new challenges," Amon told Fortune, noting that Qualcomm faced similar skepticism in 2021 when it began diversifying beyond smartphones.

Whether Qualcomm can successfully challenge Nvidia—now the world's most valuable company with a $4.7 trillion market capitalization—remains an open question. But Amon's response to doubters is characteristically direct: "A lot of people ask, 'Oh, in this crowded market, is it too late?' It's never too late for Qualcomm."

These details were first reported by Fortune.