Nvidia Wants Humanoid Robots to Work Safely Alongside Humans

Nvidia’s new humanoid robot safety platform uses AI and advanced sensors to help robots work safely alongside humans in real-world environments.

By Indrani Priyadarshini

on June 23, 2026

Nvidia is taking a major step toward making humanoid robots safer and more capable of working alongside people. To support that goal, Nvidia introduced Halos, a safety-focused software platform built using technology originally developed for self-driving vehicles. Nvidia says the system will help robots assess situations in real time, respond appropriately, and interact more naturally with people, even when physical contact is required.

Read More | How Close Are We Really to Living With Humanoid Robots?

The move comes as Nvidia and several Silicon Valley technology companies race to establish themselves in what many see as the next major growth market for artificial intelligence: robotics.

A Massive Market Opportunity

Industry forecasts suggest humanoid robots could become a significant commercial category over the next decade. According to estimates from Barclays, the global humanoid robotics market could generate as much as $200 billion in revenue by 2035.

Humanoid robots are designed to resemble humans, featuring heads, torsos, arms, and legs. Supporters of the technology believe such designs will make it easier for robots to operate in environments already built for people, including warehouses, factories, hospitals, and offices. However, widespread adoption will depend on safety.

Why Existing Safety Systems Are Not Enough

Traditional industrial robots typically operate inside restricted zones or behind safety barriers. When sensors detect a person nearby, these machines often stop or slow down to avoid accidents. According to Nvidia, that approach is insufficient for humanoid robots that are expected to move freely around people and perform collaborative tasks.

“If you think about safety in the context of a traditional robot, it is basically you need to put it in a cage or you need to have sensors which can detect there’s an obstacle and the robot comes to a stop,” said Amit Goel, Senior Director of Product Management at Nvidia. “But that is not enough for a humanoid robot.”

Read More | China Puts 100 Humanoid Robots in Real Homes: Can They Finally Handle Household Chores?

Instead, Nvidia argues that robots need the ability to understand their environment, predict potential risks, and make intelligent decisions before problems occur.

How Nvidia’s Halos Platform Works

The company plans to provide technology for robots such as Agility Robotics’ Digit, enabling them to analyse their surroundings and make independent decisions based on real-time conditions. Halos will run on Nvidia’s IGX Thor computing platform, combining safety software with specialised hardware designed for robotics applications.

Nvidia says robots powered by the system will also be able to use information from external sensors and connected infrastructure. For example, an autonomous forklift operating in a warehouse could access camera feeds positioned around corners, allowing it to detect approaching workers or vehicles before they enter its direct field of view. With that information, the robot could decide whether to continue at full speed or slow down to avoid a collision.

Expanding Beyond AI Data Centres

While Nvidia generates most of its revenue from chips used in AI data centres, the company has been steadily expanding into robotics and other AI-powered industries. The strategy is part of a broader effort to accelerate the adoption of artificial intelligence across multiple sectors. In many cases, Nvidia provides software frameworks and development tools to encourage companies to build on its technology ecosystem.

Read More | World’s first: AGIBOT brings humanoid robots into real-time electronics manufacturing

New Lab to Speed Up Robot Certification

To help bring humanoid robots to market faster, Nvidia has also established a dedicated testing facility where robotics companies can evaluate safety performance before seeking regulatory approval. The lab will allow robot manufacturers and customers to conduct safety assessments, identify potential issues, and make engineering improvements before formal certification processes begin.

Nvidia engineers will assist with pre-inspection testing and provide technical guidance to help companies meet safety requirements more efficiently. As humanoid robots move closer to real-world deployment, Nvidia is betting that advanced safety systems will be essential to earning public trust and enabling robots to work seamlessly alongside people.

News Image