Given the maturation of the smartphone market over the past decade, it has become clear that Qualcomm needs to diversify its revenue base to maintain the pace of growth to which its investors have become accustomed. Indeed, upon officially becoming CEO in 2021, one of Cristiano Amon’s first official acts was to introduce a revenue diversification strategy at its Investor Day that year and a key target market for diversification was automotive. This strategy elicited a fair amount of skepticism given the high barriers to entry and entrenched incumbent supply chain associated with the automotive market. However, Qualcomm was already quietly addressing those concerns through connectivity and cockpit solutions. Looking forward, the key to unlocking that market further is AI and they are poised to use the AI key to unlock other target markets ranging from PCs and intelligent wearable tech to advanced robotics, industrial and data centers.
Qualcomm’s venture into the automotive market began 20 years ago, well before the official introduction of its diversification strategy. Qualcomm began, as a company, developing an alternative 2G cellular technology based on its IS-95 Code Division Multiple Access (CDMA) technology, which was initially used in its OmniTRACS system for fleet communications and tracking. While Qualcomm’s focus was on handsets with the growth of the mobile market, the automotive market was never far out of the picture. With widespread adoption of mobile connectivity, Qualcomm ventured into the automotive market in the mid-1990’s with the introduction of GM’s On-Star system for mobile telematics and emergency communications. Since the mid-2000s, the growth of more advanced infotainment systems and autonomous driving platforms promoted through user expectations and the DARPA challenges has driven a need for more advanced sensor and processing platforms within the vehicles. This just happened to correlate with advancements in mobile technology. Today, sensor-based machine learning has exponentially increased the need for sensors, networking and compute resources. The new architectures that are required to fully support AI for both Advanced Driver Assistance Systems (ADAS) and infotainment, are fueling Qualcomm’s ability to penetrate this market.
Driving to Success
These market dynamics along with Qualcomm’s expertise in AI, power efficiency and high-performance computing, have allowed it to overcome other semiconductor suppliers like NXP, Renesas and Microchip upon which the automotive Tier One suppliers and OEMs traditionally relied. As such, Qualcomm now boasts a $45 billion pipeline over the next five to seven years with the pipeline consisting of approximately 67% infotainment solutions and 33% ADAS.
This year also marked the beginning of converting that pipeline to revenue with commercial launches of the Mercedes-Benz MBUX infotainment system powered by the Snapdragon Cockpit Platform and the Snapdragon Ride Platform powering the automated driving system in new BMWs starting in 2026.
In order to fully realize that pipeline and accelerate it (pun intended), ultimately, their goal, according to Qualcomm’s Vice President of ADAS Product Management, Anshuman Saxena, is to enable “human-like interactions and reactions by the car and with the car.”
To this end, Qualcomm is implementing Vision Language Action (VLA) models. These advanced neural networks orchestrate multi-modal inputs, like speech and vision, that influence the vehicle’s predictions and utilize a holistic, closed-loop feedback for the vehicle to take subsequent actions. In this way, it differs from rule-based machine learning algorithms in that it can react to rapidly changing situations as well as leverage user preferences. Multi-modal inputs include edge and cloud data, but are not limited to on-vehicle sensors, navigation input and satellite information for navigation and information over cellular networks for real-time weather and traffic information. An enabler of agentic AI, VLAs can replace the need for multiple purpose-built models to work with each other as a mixture of experts (MoE) and/or serially with one model calling upon subsequent models for specific tasks. Given the coordination involved, Qualcomm is working on developing its own VLA models while working with Tier 1 suppliers, OEMs and the robust developer community that it has garnered through its collaboration with the ecosystem.
With a system of this level of complexity, a holistic design approach is required, and this is where Qualcomm is able to fully leverage its core capabilities through its Snapdragon Ride, Flex and Cockpit platforms. Enabling VLA-based applications requires processing large amounts of data from multiple sources in real-time, but in a power efficient manner. As a result, these platforms heavily leverage Qualcomm’s expertise in power-efficient processing from its smartphone platforms.
For consumers, the VLA approach enables capabilities that mirror human interactions with the vehicle. However, it also surpasses them by providing a level of intelligence beyond the capability of the user. While humans are able to maintain situational awareness around the vehicle while driving, they do so in a serial fashion and can only reliably assess inputs in one direction and in two-dimensional space at any given time (e.g. looking out the front windshield, then the side and rearview mirrors, then the vehicle instrumentation, etc). VLA, however, can do this all simultaneously, in a 360-degree, three-dimensional sphere around the vehicle while assimilating data from sensors and other data sources the human does not have.
Is Qualcomm Succeeding In Its Revenue Diversification?
It is important to note that even though the focus is currently on Qualcomm’s revenue diversification, its core handset market is still, and will continue to be, a technology accelerator and revenue focal point for the company. Given the known headwinds of smartphone market maturation and Apple’s move away from Qualcomm, success will be determined by whether or not Qualcomm is able to offset these headwinds but also continue to deliver growth through its revenue diversification efforts.
Whether it is ultimately able to achieve this is still to be determined. However, if its recent earnings call is any indicator, Qualcomm is definitely on that path. At its fiscal Q4 2025 earnings, Qualcomm reported 10% year on year (YoY) growth driven by strength in handsets, IoT and automotive with the latter up 17% YoY and surpassing $1B quarterly revenue and ramping during a time when many of its competitors are experiencing a decrease in revenues.
This is just the beginning of Qualcomm’s road to revenue diversification. The next gear is PCs, XR, Industrial and its recent entry into data centers. In the earnings call, Qualcomm confirmed that it is on track to achieve its stated revenue targets of $8B for Auto and $14B combined for the other non-handset businesses by FY2029. The one thing all of these markets have in common is that for each, AI is a critical driving force of innovation and update cycles. If Qualcomm’s success in automotive is any indication, enabling AI, with its heavy reliance on processing performance while maintaining power efficiency will be the key that will unlock Qualcomm’s access to its other target markets and ultimately help it achieve its revenue diversification objectives.
Tirias Research tracks and consults for companies throughout the electronics ecosystem from semiconductors to systems and sensors to the cloud. Members of the Tirias Research team have consulted for IBM, Nvidia, Qualcomm, AMD and other companies throughout the data center, AI and Quantum ecosystems.
