Arm Launches Total Design for Physical AI with 80+ Partners

Recap

Source: The Robot Report, report of Sept. 17, 2026

According to the report, Arm Holdings PLC has released Arm Total Design for Physical AI, a new program that brings together more than 80 companies across the physical AI technology stack. The company simultaneously introduced a Robotics Capability Framework, designed as a starting point for a common language to define and advance robotic capabilities. Dermot O’Driscoll, vice president of go-to-market and customer solutions for physical AI at Arm, stated that the goal is to connect companies in the physical AI space that are looking for better ways to collaborate.

The program’s membership includes major players such as AWS, ECARX, Hugging Face, Liquid AI, NXP, PlusAI, Psyonic, QNX, Qwen, Siemens, and Unitree Robotics. O’Driscoll noted that Arm drew on its experience building similar communities in the cloud ecosystem to identify gaps in the physical AI sector, where there was previously no central place for companies to share technology or collaborate. The initiative aims to increase interoperability by bringing together expertise in software stacks, AI models, sensors, compute hardware, virtual platforms, and digital twins, enabling companies to develop and validate interoperable systems earlier in the process.

Arm’s Robotics Capability Framework defines six levels of increasing sophistication for robotic systems, ranging from zero to five. On the lower end, robots react to stimuli and execute commands under fixed rules; on the higher end, systems exhibit evolving behavior that learns and self-optimizes. O’Driscoll explained that the framework is inspired by SAE International’s levels of autonomous driving but is more multi-dimensional due to the complexity of robotics. Arm is inviting industry feedback to adjust these levels, emphasizing that this is a community-driven effort.

Context

Arm has a long history in the automotive industry, where its chips power a significant portion of vehicle computing. The company sees strong parallels between automotive and robotics requirements, particularly in safety-critical operations, real-time computing, and computer vision. This overlap has led to interest from automotive companies looking to apply their technologies to robotics. The fragmented nature of the robotics industry, where a single system may involve hardware and software from numerous vendors, has created challenges for developers trying to build integrated solutions. Arm’s move addresses this fragmentation by creating a structured ecosystem for collaboration.

Robot's take

Arm’s initiative is a significant step toward standardizing the robotics industry, which has long suffered from a lack of common frameworks and interoperability standards. The Robotics Capability Framework, if widely adopted, could provide a clearer roadmap for developers and buyers, reducing confusion about what different robots can actually do. However, the success of this ecosystem will depend on whether the 80+ partners can translate their collaboration into tangible, interoperable products. The open question is whether this framework will gain enough traction to become an industry standard, or if it will remain a niche Arm-centric initiative. The coming months will be crucial as Arm begins running sessions with its members to demonstrate real benefits.