NVIDIA Targets $200B Humanoid Robot Market

According to industry reports, AI chip leader NVIDIA is accelerating its push into humanoid robots, positioning safety as the key threshold for large-scale commercial deployment. On Monday, the company announced a new integrated hardware and software stack designed specifically for humanoid robotics, aiming to enable machines to make real-time safety decisions even during close-range human interaction or physical contact. The goal is to expand deployment across factories, logistics, retail, and healthcare environments.

Market expectations for humanoid robots continue to rise, with the sector widely viewed as the next major growth wave following generative AI. Barclays estimates that the humanoid robotics industry could reach a revenue scale of around US$200 billion by 2035. Major Silicon Valley technology companies, including NVIDIA, are increasingly investing in core technologies to capture opportunities in what could become a multi-billion-unit global robotics market.

Expanding from autonomous driving to robotics safety platforms

NVIDIA’s latest solution, the Halos platform, builds on safety technologies originally developed for autonomous driving systems. The company emphasizes that future humanoid robots must achieve human-level environmental awareness in order to safely collaborate with people in real-world operational settings.

Today’s industrial robots typically rely on conservative safety mechanisms. When sensors detect nearby humans or obstacles, machines will slow down or stop completely to avoid collision risks. While effective for safety, this approach limits productivity and restricts more complex human–robot collaboration tasks, such as handing over objects or assisting with heavy lifting.

According to NVIDIA senior director of product management Amit Goel, traditional robot safety models are often based on physical isolation or immediate shutdown upon obstacle detection. However, he noted that such methods are no longer sufficient for the complexity of humanoid robotics operating in dynamic environments.

Through the combination of the Halos software framework and the IGX Thor computing platform, NVIDIA aims to enable robots to interpret their surroundings in real time, allowing them to determine which objects can be safely touched, moved, or handled with force—rather than simply stopping operation as a default safety response.

Toward scalable deployment in industrial and commercial environments

NVIDIA stated that the Halos platform will initially support humanoid robot systems such as Digit developed by Agility Robotics. These systems are expected to integrate not only onboard sensors but also external cameras and environmental monitoring infrastructure to obtain a more complete understanding of surrounding conditions.

For example, in warehouse automation scenarios, autonomous forklifts could use external camera feeds to see around corners and assess whether to maintain speed or slow down, improving both operational efficiency and workplace safety.

To accelerate commercialization, NVIDIA has also established dedicated testing facilities to assist robotics developers with safety validation and certification preparation. The company’s engineering teams will conduct pre-submission assessments and technical adjustments before regulatory review, helping shorten time-to-market for new robotic systems.

Agility Robotics Chief Technology Officer Pras Velagapudi highlighted a fundamental difference between humanoid robots and autonomous vehicles. While self-driving cars primarily focus on avoiding collisions, humanoid robots must also understand when and how to interact physically with objects—such as lifting, carrying, or supporting loads. This requires significantly more advanced safety intelligence.

Industry development is expected to begin in highly structured environments such as warehousing and logistics, before gradually expanding into retail, healthcare, and construction applications.

As AI technologies continue to advance rapidly, NVIDIA is not only maintaining its leadership in AI data center GPUs but also strategically expanding into robotics. Analysts suggest that if humanoid robots can successfully overcome safety and regulatory challenges, the sector could become a major new engine driving the next phase of AI industry growth.