Qualcomm is set to present an expanded strategy beyond its core smartphone business at an upcoming investor day, where executives are expected to provide more detail on the company’s push into AI data center chips. Analysts anticipate announcements of new customers as Qualcomm seeks to establish a presence in a market currently dominated by Nvidia.
The move comes as the smartphone segment that has been a major source of revenue for Qualcomm faces intensifying pressure. A global memory chip shortage - driven in part by surging demand for AI infrastructure - has constrained that market, while large handset makers including Apple and Samsung are increasingly designing chips internally rather than relying on external suppliers.
In response to these headwinds, Qualcomm has been broadening its business mix beyond handsets, extending efforts into automotive and data center markets. The company has tried multiple times to grow its data-center business and is now re-entering a fast-growing AI chip market that several observers describe as hyper-competitive.
Bank of America analysts flag the competitive landscape as crowded, citing established players such as Nvidia and newer entrants like Cerebras, alongside a range of custom chip options available from hyperscalers and cloud providers - including Amazon’s Graviton and Google’s Axion. Those analysts caution that the environment will be challenging as Qualcomm seeks to penetrate a space with entrenched incumbents and diverse custom solutions.
Qualcomm said in April that it intends to begin shipping processors and other AI chips designed for data center workloads by year-end. The company indicated it is collaborating with customers on three categories of devices: central processing units, inference accelerators, and custom application-specific integrated circuits (ASICs). The custom ASIC segment has been expanding for competitors such as Broadcom and Marvell.
AI inference - the execution of trained models in production - has emerged as an important battleground in the market for data-center chips. Bank of America analysts estimate Qualcomm’s push into data-center products could generate roughly $2 billion to $5 billion in annual revenue by fiscal 2027-2028, describing that as modest relative to the broader market.
Investors following the event are expected to look for updated long-term financial targets, including explicit growth objectives for the company’s non-handset businesses. Market attention is also likely to center on Qualcomm’s recently announced $4 billion all-stock acquisition of AI software startup Modular, disclosed earlier on the same day. That deal positions Qualcomm as a competitor to Nvidia’s proprietary CUDA software ecosystem, which the company notes has secured a substantial developer base.
Key context and takeaways
- Qualcomm plans to disclose more on its AI data-center strategy and may name new customers at investor day.
- The company is targeting three chip categories - CPUs, inference accelerators and custom ASICs - with shipments for data centers slated to start by year-end.
- Analysts warn the AI data-center chip market is highly competitive and dominated by incumbents and custom-chip offerings from cloud providers.
Sectors impacted: Semiconductors, cloud/data-center infrastructure, smartphone hardware, automotive electronics.