Skild AI Unveils Groundbreaking 'Skild Brain' Model for Multi-Purpose Robots

Amazon-backed Skild AI unveils general-purpose AI model for multi-purpose robots

July 29 (Reuters) - Robotics startup Skild AI, backed by Amazon.com and Japan's SoftBank Group (9984.T), opens new tab, on Tuesday unveiled a foundational artificial intelligence model designed to run on nearly any robot -- from assembly-line machines to humanoids. The model, called Skild Brain, enables robots to think, navigate and respond more like humans. Its launch comes amid a broader push to build humanoid robots capable of more diverse tasks than the single-purpose machines currently found on factory floors. In demonstration videos, Skild-powered robots were shown climbing stairs, maintaining balance after being shoved, and picking up objects in cluttered environments - tasks that require spatial reasoning and the ability to adapt to changing surroundings. The company said its model includes built-in power limits to prevent robots from applying unsafe force. Skild trains its model on simulated episodes and human-action videos, then fine-tunes it using data from every robot running the system. Co-founders Deepak Pathak and Abhinav Gupta told Reuters in an exclusive interview that the approach helps tackle a data scarcity problem unique to robotics. "Unlike language or vision, there is no data for robotics on the internet. So you cannot just go and apply these generative AI techniques," Pathak, who serves as CEO, told Reuters. Robots deployed by customers feed data back into Skild Brain to sharpen its skills, creating the same "shared brain," said Gupta, who previously founded Meta Platforms' (META.O), opens new tab robotics lab in Pittsburgh. Skild's clients include LG CNS - the IT solutions arm of LG Group - and other unnamed partners in logistics and other industrial applications. Unlike software, which can scale quickly, robotics requires physical deployment, which takes time, but Skild's approach allows robots to add new capabilities across different industries quickly, said Raviraj Jain, partner at the startup's investor Lightspeed Venture Partners. The two-year-old startup, which has hired staff from Tesla, Nvidia and Meta, raised $300 million in a Series A funding round last year that valued it at $1.5 billion. Its investors include Menlo Ventures, Khosla Ventures, Sequoia Capital and Amazon.com (AMZN.O), opens new tab founder Jeff Bezos. Reporting by Akash Sriram in Bengaluru; Editing by Tasim Zahid Our Standards: The Thomson Reuters Trust Principles., opens new tab * Suggested Topics: * Artificial Intelligence Akash Sriram Thomson Reuters Akash reports on technology companies in the United States, electric vehicle companies, and the space industry. His reporting usually appears in the Autos & Transportation and Technology sections. He has a postgraduate degree in Conflict, Development, and Security from the University of Leeds. Akash's interests include music, football (soccer), and Formula 1.

Amazon-backed firm unveils shared brains for all types of robots

Skild Brain allows robots to think, navigate, and react more like humans. Its debut aligns with the growing drive to create humanoid robots that can handle far more varied tasks than today's single-purpose factory machines. The model can work on various quadrupeds, humanoids, table-top arms, mobile manipulators, and more. In the demos released by Skild AI, robots powered by the model were seen climbing stairs, staying balanced after being pushed, kicked and picking up objects in cluttered spaces - tasks that demand spatial awareness and adaptability to changing environments. The company noted that the model has built-in power limits to prevent robots from exerting unsafe force. Skild trains the system using simulated scenarios and human-action videos, fine-tuning it with real-world data collected from every robot running the model. One of the biggest hurdles in creating a true robotics foundational model is the lack of a large data set for robotics. Gathering such data in the real world is both slow and prohibitively expensive. Many researchers use the vision-language models (VLMs) - trained on massive image and text collections - and add a tiny fraction of real-world robot data, often less than one percent.

Watch This Unicorn Company's Humanoid Robots Complete Helpful Tasks

Robotics startup Skild AI, backed by Amazon.com and Japan's SoftBank Group, on Tuesday unveiled a foundational artificial intelligence model designed to run on nearly any robot -- from assembly-line machines to humanoids. The model, called Skild Brain, enables robots to think, navigate and respond more like humans. Its launch comes amid a broader push to build humanoid robots capable of more diverse tasks than the single-purpose machines currently found on factory floors. In demonstration videos, Skild-powered robots were shown climbing stairs, maintaining balance after being shoved, and picking up objects in cluttered environments - tasks that require spatial reasoning and the ability to adapt to changing surroundings. The company said its model includes built-in power limits to prevent robots from applying unsafe force. Skild trains its model on simulated episodes and human-action videos, then fine-tunes it using data from every robot running the system. Co-founders Deepak Pathak and Abhinav Gupta told Reuters in an exclusive interview that the approach helps tackle a data scarcity problem unique to robotics. "Unlike language or vision, there is no data for robotics on the internet. So you cannot just go and apply these generative AI techniques," Pathak, who serves as CEO, told Reuters. Robots deployed by customers feed data back into Skild Brain to sharpen its skills, creating the same "shared brain," said Gupta, who previously founded Meta Platforms' robotics lab in Pittsburgh. Skild's clients include LG CNS - the IT solutions arm of LG Group - and other unnamed partners in logistics and other industrial applications. Unlike software, which can scale quickly, robotics requires physical deployment, which takes time, but Skild's approach allows robots to add new capabilities across different industries quickly, said Raviraj Jain, partner at the startup's investor Lightspeed Venture Partners. The two-year-old startup, which has hired staff from Tesla, Nvidia and Meta, raised $300 million in a Series A funding round last year that valued it at $1.5 billion. Its investors include Menlo Ventures, Khosla Ventures, Sequoia Capital and Amazon.com founder Jeff Bezos. Reporting by Akash Sriram in Bengaluru; Editing by Tasim Zahid The final deadline for the 2025 Inc. Power Partner Awards is Friday, August 8, at 11:59 p.m. PT. Apply now.

Amazon-backed Skild AI unveils general-purpose AI model for multi-purpose robots - The Economic Times

The model, called Skild Brain, enables robots to think, navigate and respond more like humans. Its launch comes amid a broader push to build humanoid robots capable of more diverse tasks than the single-purpose machines currently found on factory floors.Robotics startup Skild AI, backed by Amazon and Japan's SoftBank Group, on Tuesday unveiled a foundational artificial intelligence model designed to run on nearly any robot - from assembly-line machines to humanoids. The model, called Skild Brain, enables robots to think, navigate and respond more like humans. Its launch comes amid a broader push to build humanoid robots capable of more diverse tasks than the single-purpose machines currently found on factory floors. In demonstration videos, Skild-powered robots were shown climbing stairs, maintaining balance after being shoved, and picking up objects in cluttered environments - tasks that require spatial reasoning and the ability to adapt to changing surroundings. The company said its model includes built-in power limits to prevent robots from applying unsafe force. Skild trains its model on simulated episodes and human-action videos, then fine-tunes it using data from every robot running the system. Co-founders Deepak Pathak and Abhinav Gupta told Reuters in an exclusive interview that the approach helps tackle a data scarcity problem unique to robotics. "Unlike language or vision, there is no data for robotics on the internet. So you cannot just go and apply these generative AI techniques," Pathak, who serves as CEO, told Reuters. Robots deployed by customers feed data back into Skild Brain to sharpen its skills, creating the same "shared brain," said Gupta, who previously founded Meta Platforms' robotics lab in Pittsburgh. Skild's clients include LG CNS - the IT solutions arm of LG Group - and other unnamed partners in logistics and other industrial applications. Unlike software, which can scale quickly, robotics requires physical deployment, which takes time, but Skild's approach allows robots to add new capabilities across different industries quickly, said Raviraj Jain, partner at the startup's investor Lightspeed Venture Partners. The two-year-old startup, which has hired staff from Tesla, Nvidia and Meta, raised $300 million in a Series A funding round last year that valued it at $1.5 billion. Its investors include Menlo Ventures, Khosla Ventures, Sequoia Capital and Amazon founder Jeff Bezos.