Nvidia is staking a claim in the next chapter of AI by moving from data centers into what it calls Physical AI, building safety and compute layers for robots that walk, lift, and work alongside people. Its Halos for Robotics initiative bundles hardware, software, and safety systems, and early partners like Agility are already testing the stack in real industrial settings. Analysts are increasingly bullish on the larger market for humanoid and industrial robots, and Nvidia’s recent financials and capital returns give the company fuel to lead that shift.
Nvidia (NVDA) is no longer just the engine behind cloud and server AI. The company says the next major wave will be Physical AI, where machines move through and interact with the real world in ways that require safe, reliable compute. That shift demands a different kind of platform—one that combines sensors, simulation, runtime safety, and high-performance inference near the edge.
Halos for Robotics is Nvidia’s attempt to answer that need with a full-stack open safety system for physical AI and robotics. The platform ties together AI compute and sensor bridges with what the company calls Halos OS to create a safety-aware operating environment. Nvidia positions this not as a single product but as an ecosystem designed to reduce the friction of deploying robots where humans work and live.
Agility is notably the first public integrator to adopt Halos for Robotics, using it to add safety functions to its Digit humanoid. “For Digit, NVIDIA IGX Thor delivers industrial-grade AI compute with built-in safety capabilities, while Halos Core supports the software layer for safety-related operating functions.” That exact pairing shows Nvidia is selling more than chips; it is packaging compute, middleware, and safety under one banner.
>
“Humanoid robotics is really on an upward trajectory. The size of the market today is really small; it’s 2 to 3 billion [dollars], but we see it going up to $200 billion in 2035,” as also quoted by CNBC.
Big-name analysts are already sketching out enormous potential. Wedbush’s Dan Ives calls humanoid robots one of the biggest opportunities in the AI boom and says they “will change the way consumers and businesses operate over time.” Other firms offer even larger forecasts, from Morgan Stanley’s multi-trillion outlook by mid-century to Bank of America and Goldman Sachs predicting massive shipment and addressable-market growth over the next decade.
That market story matters for Nvidia because scaling real-world robots depends on compute, simulation, and safety layers as much as the physical designs. If factories, warehouses, and service environments adopt humanoids or advanced mobile robots, the need for high-throughput AI inference, simulation tooling, and certified safety middleware could drive sustained demand for Nvidia’s stack.
Nvidia’s recent financials back up its market positioning: the company posted strong earnings with data center revenue surging and management guiding to an aggressive revenue target that outpaced analyst expectations. The firm also boosted buybacks and raised its dividend, moves that both reward shareholders and give Nvidia leverage to invest in partnerships, software development, and ecosystem growth tied to robotics and safety.
Analyst sentiment is overwhelmingly positive, reflecting both the AI tailwind and the robotics opportunity. “Of the 49 analysts covering NVDA stock, 43 analysts have a “Strong Buy,” three have a “Moderate Buy,” two analysts offer a “Hold” rating, and one has a “Strong Sell.” The mean target price of $303.71 implies roughly 52% potential upside from current levels.” If robotics adoption even partially matches those projections, demand for Nvidia’s compute and safety infrastructure could become a major new growth vector.
