Qualcomm first signaled its data-center ambitions in June, unveiling the Dragonfly C1000 CPU and saying Meta would use the processor when production begins in 2028. The company pitched the chip for agentic AI, where power efficiency matters as much as raw compute, and set a target of $15 billion in data-center revenue for fiscal 2029.¹

That expansion now has a far larger ally. Qualcomm and Amazon Web Services announced a multigenerational collaboration on customized silicon for AWS AI infrastructure, with a particular focus on inference — the work of running models after they have been trained. The companies are also developing “high-performance optical connectivity solutions,” while Qualcomm will use Amazon’s AI servers to accelerate its own chip-design process.²

The companies’ shared argument is that AI’s growth demands a different infrastructure balance. Their statement cited “unprecedented demand” for compute, storage, networking, memory bandwidth and energy-efficient systems, pairing AWS’s infrastructure with Qualcomm’s power-efficient processing and silicon-design expertise.¹

For Qualcomm, the commercial terms underscore both the opportunity and the pressure. It issued Amazon warrants for 25 million shares at $161.26 apiece — up to $4 billion — which vest in stages linked partly to Amazon buying as much as $60 billion of Qualcomm server chips and related technology. Qualcomm shares rose 4% after the announcement.¹

The deal does not dislodge Nvidia overnight. Nvidia remains the defining force in AI accelerators, while AMD and Intel are also chasing rising demand for data-center CPUs. But AWS’s backing gives Qualcomm a consequential validation: a hyperscaler is willing to help test whether efficient CPUs, tailored silicon and faster interconnects can become a credible alternative in the AI buildout.