FieldAI, a leader in robotic autonomy and embodied AI, has secured $405 million through two consecutive funding rounds backed by investors including Bezos Expeditions, BHP Ventures, Canaan Partners, Emerson Collective, NVentures, the venture arm of NVIDIA, Temasek, Prysm, Intel Capital, and Khosla Ventures, among others. Previous supporters such as Gates Frontier and Samsung also participated. The rounds were oversubscribed, reflecting rapid adoption of FieldAI’s general-purpose robotics platform, which has already been deployed successfully in hundreds of complex industrial environments worldwide.

The company is pioneering a single “software brain” capable of powering multiple types of robots across diverse settings. From Japan to Europe to the U.S., FieldAI systems are already operating daily in industries such as construction, energy, manufacturing, logistics, and inspection. Its models make decisions in real time at the edge, seamlessly integrating into customer workflows. With significant operational hours and real-world data logged, FieldAI has demonstrated continuous improvement, scalable autonomy, and meaningful cost efficiency. Rising demand for automation to address labor gaps, safety challenges, and productivity goals is accelerating the adoption of its technology.

With ambitions to double the staff by the end of the year, the new funding will be utilized to support strategic hiring, propel improvements in locomotion and manipulation, and broaden FieldAI’s global presence. According to Vinod Khosla of Khosla Ventures, FieldAI’s unique approach to embodied intelligence provides a pragmatic and scalable path to autonomy, unlocking long-term economic and social value.

At the heart of FieldAI’s platform are Field Foundation Models (FFMs), a new generation of “physics-first” foundation models designed specifically for robotics. Unlike vision or language models adapted for physical use, FFMs are built from the ground up to account for risk, uncertainty, and real-world physical constraints. This design allows robots to operate safely in unpredictable conditions, navigating environments without maps, GPS, or predefined paths. FieldAI CEO Ali Agha explained that the company’s years of field research informed the development of intrinsically risk-aware architectures, avoiding the limitations of retrofitted AI approaches.

FFMs enable robots to dynamically adapt to new challenges without reprogramming and have already been validated across quadrupeds, humanoids, wheeled machines, and passenger-scale vehicles. Their hardware-agnostic nature means that a wide variety of robots can run on the same intelligence core, accelerating deployment and scaling.

FieldAI’s team includes veterans from DeepMind, Google Brain, Tesla Autopilot, NASA JPL, SpaceX, Zoox, Cruise, DARPA, TRI, and more—bringing together deep research expertise and proven deployment at scale. Headquartered in Irvine, California, the company is driving a new era of autonomous robotics, equipping industries worldwide to scale automation safely and effectively.