Qualcomm’s robotics chip push turns physical AI into the next growth narrative beyond phones

date
09:09 04/03/2026
avatar
GMT Eight
Qualcomm is trying to turn robotics into a repeat of its smartphone playbook: be the power-efficient compute platform that scales across OEMs, while partners differentiate on software, sensors, and form factor. At CES 2026, the company introduced a comprehensive-stack robotics architecture and unveiled its Dragonwing IQ10 Series processor aimed at industrial autonomous mobile robots and full-size humanoids, positioning it as an energy-efficient brain of the robot for real-world deployment.

Qualcomm’s bet is that robotics is now ready for an edge-AI inflection: perception, motion planning, and on-device decisioning are becoming practical as models and sensor fusion improve, but they still require low-latency compute that can run untethered inside machines. In its February earnings call, Qualcomm said it had formally announced an expansion into advanced robotics and described its general-purpose robotics architecture as supporting advanced perception and motion planning using models such as VLAs and VLMs, enabling robots to perceive, reason, adapt and act in real-world environments, with IQ10 designed to accelerate commercialization across household, industrial, and humanoid robots. The company explicitly framed this as leveraging strengths it developed in autonomy and safety-grade silicon, arguing many of the same drivers behind its automotive leadership apply to advanced robotics as well.

The ecosystem angle is central to how Qualcomm thinks it wins. In the CES announcement distributed via Business Wire, Qualcomm said it is building a partner network and called out collaborations or engagements with firms including Figure and Kuka Robotics, alongside a broader list of robotics and embedded partners, while describing IQ10 as the latest addition to its premium-tier robotics processors for humanoids and advanced AMRs. This horizontal platform approach is intended to reduce the integration burden for robotics OEMs by bundling compute, connectivity, and a software/tooling stack that can be reused across deployments, rather than treating each robot program as a one-off engineering effort.

For investors, the robotics story is also a portfolio story: Qualcomm is reinforcing that its next chapter is multi-engine growth across automotive and IoT, not just handset cycles. In the same earnings call, Qualcomm noted near-term handset guidance was being impacted by memory availability dynamics tied to DRAM supply and HBM prioritization for data centers, while emphasizing continued momentum across multiple vectors including robotics and industrial IoT. In its Q1 FY26 earnings presentation, Qualcomm highlighted its formal expansion into advanced robotics (including Dragonwing IQ10) alongside strong Automotive and IoT traction, presenting robotics as part of a broader edge-AI push.