OpenMind Is Building the AI Brain for Humanoid Robots

Humanoid robots can look ready for the future in a polished demo.

Then Jan Liphardt bought one, put it in his home, and discovered that it could barely do anything. It could stand, but it could not become the R2-D2-like helper, teacher, companion, or coworker he had imagined.

In this episode of Automated, Brian Heater speaks with the OpenMind founder and Stanford bioengineering professor about the enormous gap between a humanoid that works in a lab and one that can safely operate around people in the real world.

They discuss why the last 3% of deployment can take nearly all the work, why a socially intelligent robot needs body language as well as speech, and how OpenMind uses natural language to make robot decisions easier for people to inspect, debug, and trust.

Brian and Jan also explore why OpenMind is not trying to solve hands or high-speed control, how OM1 and FABRIC fit into its approach, why there is no single robot data problem, and when physics may teach a robot more than videos of humans.

The conversation also gets into Jan’s unconventional approach to hiring, the value of bringing fresh eyes to robotics, and why “humanoid brain architect” is now a real job.

If you want a grounded look at what today’s humanoid robots can actually do, what the demos leave out, and what it will take to turn promising hardware into useful machines, this is the conversation.

KEY MOMENTS
00:00 The humanoid Jan bought could not do anything
01:36 Introducing Jan Liphardt and OpenMind
03:36 Reed College, science culture, and its nuclear reactor
06:47 The blue glow every science nerd wants to see
07:09 The hardware and software thread through Jan’s career
10:20 How Jan identifies unconventional engineering talent
13:15 Why elite degrees do not guarantee great engineers
15:56 What the three-year and six-year PhD models reveal
20:09 Why Jan combines academia with the real world
21:51 The strange incentives inside academic research
24:16 How Stanford Biodesign changed his approach to startups
26:49 The humanoid delivery that sparked OpenMind
28:50 Why the last 3% of humanoid deployment is so hard
31:01 “Humanoid brain architect” is now a real job
32:13 The robotics problems OpenMind deliberately avoids
37:01 Why social intelligence requires more than speech
39:10 The incredible value of ignorance in robotics
40:17 What families actually want from a home robot
42:33 Making robot decisions understandable and trustworthy
46:20 How OM1 and FABRIC fit into OpenMind’s stack
47:03 Why there is no single robot data problem
50:52 What Jan learned from Steve Cousins and Willow Garage

Connect with Jan Liphardt
https://www.linkedin.com/in/jan-liphardt

Learn more about OpenMind
https://openmind.com/

Learn more about Jan’s work at Stanford
https://profiles.stanford.edu/jan-liphardt

Register for the Automated Happy Hour in Mountain View on September 22
https://luma.com/n688dcl5

Learn more about the Advanced Vision and AI Conference on September 23 and 24
https://www.automate.org/events/advanced-vision-and-ai-conference/register

We’d love to hear from you.
Have thoughts or guest suggestions?
Reach us at podcast@automate.org

You can find the transcript and more episodes of Automated at automated.fm

Unlock full access to Automated and explore everything automation.
Subscribe today and leave a review on YouTube, Apple Podcasts, Spotify, and the Automated Newsletter.
https://www.youtube.com/@automatedpodcast
https://podcasts.apple.com/us/podcast/automated-with-brian-heater/id1837762221
https://open.spotify.com/show/60olq6brlBEIJWggx2fMR6
https://www.automate.org/automation/automated-newsletter

You can also find us on:
LinkedIn https://www.linkedin.com/showcase/automated-podcast-by-a3/
Instagram https://www.instagram.com/automatedpod/