While many companies focus on robot hardware—like arms, legs, or grippers—OpenMind is building something deeper: the software that powers them. Based in Silicon Valley, OpenMind has developed OM1, a software layer that acts as an operating system for humanoid robots. The company compares itself to Android in the mobile world—open, flexible, and compatible with any hardware.
Founded in 2024 by Stanford professor Jan Liphardt, OpenMind believes the future of robotics lies in intelligent interaction. Robots have long performed repetitive tasks. However, as they move into homes and workplaces, they need smarter software. Therefore, they require a system that thinks more like a human.
“All of a sudden, this world is opening where machines are able to interact with humans in ways I’ve certainly never before seen,” Liphardt told TechCrunch. “We’re very much believers here that it’s not just about the humans. We really think of ourselves as a company that is a collaboration between machines and humans.”
To enable deeper collaboration, OpenMind unveiled a new protocol called FABRIC. This system allows robots to verify each other’s identity. They can also share context, data, and learned behaviors in real time. Unlike humans, machines can learn almost instantly. So, giving them a way to connect means they can train faster. They can absorb new skills without human input.
For example, one robot could learn a new language and share it with others. As a result, robots across regions could communicate with more people. This would happen far faster than if each had to be taught individually. “Humans take it for granted that they can interact with any other human on Earth,” Liphardt said. “We’ve built infrastructure to trust, call, text, and coordinate. Machines will need the same.”
OpenMind is preparing to ship its first fleet of 10 robotic dogs powered by OM1 by September. These quadrupeds will be placed in homes and labs. Their goal is to gather real-world feedback. Liphardt expects users to return with criticism and requests. “We full well expect all the humans hosting these quadrupeds will come back with a long list of things they didn’t like or want,” he said. “It’s up to us to very quickly iterate and improve.”
The company recently raised $20 million in a funding round. Pantera Capital led the round, with participation from Ribbit, Coinbase Ventures, Pebblebed, and several angel investors. The capital will support product development and scaling the OM1 platform.
Now, OpenMind’s top priority is getting robots into the real world. Feedback from early users will help align robot capabilities with actual human needs. “Our goal is to do as many of these tests as we can,” Liphardt said. “That way, we can rapidly identify the most interesting opportunities where robots and human needs match.”
Because robots are only as useful as their software, OpenMind focuses on the brain, not the body. By building a universal OS, it allows different robot makers to innovate faster. Eventually, OM1 could become the standard for all humanoids—no matter the manufacturer.
Moreover, the FABRIC protocol opens new possibilities. Robots could share safety data, navigation maps, or even emotional responses. Over time, this network effect could lead to a shared intelligence layer.
While competitors build isolated systems, OpenMind bets on openness. Its hardware-agnostic model means OM1 can run on any robot. So, developers don’t need to reinvent the OS for every new design.
In short, OpenMind isn’t just building software. It’s building a future where robots learn from each other. Where they work alongside humans—not just for them.
READ: Big Tech’s AI Spending Soars as Investors Cheer Returns
