Nvidia is taking artificial intelligence (AI) to the next level with the launch of Cosmos, its latest family of basic AI models. Announced during CEO Jensen Huang’s keynote speech at CES 2025 in Las Vegas, Cosmos, he says, is specifically designed to teach robots and machines how to interact with the physical world. Cosmos was trained on 20 million hours of real footage of “humans walking, moving hands, manipulating things,” Jensen said. From industrial robots to humanoids and self-driving cars, the applications of this technology are vast and fascinating.

What sets Cosmos apart?

While language models like ChatGPT or Bard learn to generate text by digesting heaps of written content, Cosmos focuses on the visual and physical domain. Huang explained that the AI ​​model was trained on 20 million hours of real footage of humans performing everyday actions – walking, moving hands, manipulating objects. “It’s not about making art,” Huang said. “It’s about teaching AI to understand the physical world.”

During the presentation, Huang clearly showed Cosmos in action, simulating a warehouse environment where boxes fell off the shelves. This type of realistic video generation is important for training robots to recognize and respond to real-world scenarios such as workplace accidents. Companies can enhance Cosmos by tailoring it to specific needs using their proprietary data.

The impact of Cosmos is already being felt across industries. Humanoid robot startups like Agility Robotics and Figure AI are leveraging the technology, along with self-driving car companies like Uber, Wabi, and Wave. These businesses hope to push the limits of what robots can achieve, whether it’s navigating complex environments or performing complex tasks.

isaac complements the universe

To complement Cosmos, Nvidia also unveiled an upgrade to its Isaac robot simulation platform. The new feature allows robots to learn new tasks with unprecedented efficiency. For example, builders can use a few examples of a desired action, such as grasping an object, and Isaac will generate a multitude of synthetic training data to perfect the skill.

Huang’s CES keynote speech included life-size digital images of 14 humanoid robots from industry leaders such as Tesla, Boston Dynamics and Agility Robotics, emphasizing Nvidia’s ambition to shape the future of robotics. “With Cosmos and Isaac, we are giving developers the tools to build robots that can truly understand and operate in our world,” Huang said.

Nvidia’s bold vision underlines its leadership in the AI ​​field, combining innovation with practicality. The integration of AI into robotics could redefine industries from logistics and manufacturing to autonomous driving. With Cosmos, Nvidia is paving the way for machines to become more human – not in thought, but in motion.