Robotics Foundation Models Poised for $150B Market, Driving Physical AI Adoption Across Industries

A new report from global technology intelligence firm ABI Research forecasts that robotics foundation models will drive a global market worth US$150 billion by 2036. This significant projection underscores a pivotal transformation in the automation landscape, as advances in what's termed 'Physical AI' enable robots to execute increasingly complex and adaptable tasks that have historically resisted automation. The report emphasizes that this growth will extend beyond traditional industrial robotics, opening vast opportunities across diverse sectors including manufacturing, warehousing, logistics, healthcare, retail, hospitality, and restaurants. This development is critical for cloud and DevOps practitioners because it signals a massive expansion of intelligent edge computing and the need for robust MLOps pipelines tailored for physical systems. The ability of robots to learn, reason, and manipulate varied and deformable objects with unprecedented robustness means that the demand for scalable AI infrastructure, both in the cloud for training and at the edge for inference, will skyrocket. For organizations looking to leverage automation, this isn't just about deploying more robots; it's about integrating highly intelligent, adaptable systems that require continuous data feedback loops and model updates, directly impacting how cloud resources are consumed and managed. The shift from fixed-function robots to adaptable, AI-driven systems fundamentally changes operational paradigms, requiring a more dynamic and intelligent infrastructure. This trend fits squarely within the broader, well-established convergence of AI, cloud computing, and edge deployment. The evolution of large language models (LLMs) into multimodal foundation models has already demonstrated the power of generalized AI. Now, this paradigm is extending into the physical world, creating 'Physical AI' where robots leverage these models for enhanced perception, decision-making, and manipulation. Cloud providers like AWS, Microsoft Azure, and Google Cloud are already positioning themselves to support the extensive model training, simulation, and lifecycle management required for these advanced robotic systems. Concurrently, hardware innovators such as NVIDIA, with its Jetson platform, are leading in on-device robotics compute, while competitors like Intel, AMD, Qualcomm, and Ambarella are developing alternative solutions. This mirrors the broader trend of AI workloads shifting between centralized cloud training and distributed edge inference, a pattern seen across autonomous vehicles, smart factories, and intelligent IoT. In practice, this means practitioners should focus on developing expertise in MLOps for edge devices, understanding the nuances of deploying and managing AI models on resource-constrained hardware, and ensuring seamless integration with cloud-based training and data management platforms. Organizations should also evaluate their cloud strategies to ensure they can accommodate the compute-intensive demands of robotics simulation and model refinement. Furthermore, the report highlights manufacturing as the leading near-term opportunity, projected to reach a US$30 billion market by 2036, followed by warehousing and logistics at US$21 billion, and healthcare at US$16 billion. This indicates where early adopters and significant investment will be concentrated, guiding where practitioners might find the most immediate impact and career opportunities. The emphasis on interoperability, as seen in other recent developments in warehouse automation, will also be crucial, requiring open standards and collaborative ecosystems to realize the full potential of these advanced robotic systems.