Nvidia launched Halos for Robotics in June 2026, extending a full-stack safety architecture originally built for autonomous vehicles to warehouse robots, in-factory humanoids, and surgical machines.

The original Halos system debuted in 2025, bundling hardware and software tools that help developers implement guardrails in self-driving cars and other autonomous vehicles. The June 2026 expansion marks the first time Nvidia has positioned a single safety framework across the full spectrum of what it calls physical AI — robotics and self-driving technology that the company has backed with billions of dollars in investment.

That bet sits alongside the AI data center business that drives Nvidia's multi-trillion-dollar market capitalization. Investors may be pricing in GPU demand, but Nvidia has spent years building a parallel stack for machines that move.

Why safety, and why now?

Nvidia frames the problem as a sequencing issue: capabilities have arrived faster than the safety infrastructure needed to deploy them.

"Now the AI models are getting capable, the robot hardware is getting capable, and a thing we thought is going to be the next bottleneck is safety," Amit Goel, head of robotics ecosystem and edge computing at Nvidia, told Ars Technica. "So that's why we launched our Halos for Robotics to unlock the capability of these systems."

The logic is commercial as much as technical. Autonomous mobile robots in warehouses, humanoid robots walking factory floors, and surgical robots all face deployment barriers that are regulatory and liability-driven, not purely performance-driven. A safety system that developers can build on — rather than reinvent per product — could shorten the path from pilot to production.

What Halos for Robotics covers?

According to Nvidia, Halos for Robotics is the industry's first full-stack safety system for physical AI. It targets three deployment categories:

• Autonomous mobile robots operating in warehouses and logistics environments
• Humanoid robots working inside factories among people
• Surgical robots, where safety failures carry the highest stakes

The original automotive Halos, launched in 2025, provided the template: hardware and software tools that let developers implement guardrails rather than build safety mechanisms from scratch.

The stakes for Nvidia

Nvidia's physical AI push is a bet that the next wave of AI compute demand comes from machines embedded in the physical world, not just data centers. The company has invested billions in robotics and self-driving technology while its data center business fuels a market cap measured in trillions.

Safety is the gating factor. If robots and autonomous vehicles cannot be certified and trusted in shared spaces with people, the deployment volumes — and the silicon they would consume — stay limited. By supplying the safety layer itself, Nvidia positions its platform at the point where deployment decisions get made.

Goel's framing suggests Nvidia sees capability and safety as the last two dominoes: the models and hardware are ready, and the company intends to remove safety as the remaining constraint. Whether Halos becomes the de facto safety foundation for physical AI — or one framework among many — will shape how quickly robotaxis, warehouse fleets, and humanoids reach commercial scale.