Why the Next Great Home Robot Will Not Be Humanoid
Automated Podcast 50:45
779 views · 21 likes Watch on YouTube ↗
Home robots have been promised for years.
So why did so many of them fail?
In this episode of Automated, Colin Angle explains why earlier generations of home robotics hit a wall, and why AI may finally change that.
After co-founding iRobot and helping bring Roomba into millions of homes, Colin is back with Familiar Machines & Magic, a new company focused on what he calls consumer physical AI.
He explains why the hardest problem in robotics was never just movement. It was perception, context, and the ability to understand what is happening in the real world.
The conversation also explores one of the biggest lessons from Roomba: even a technically impressive robot can fail if it does not fit naturally into daily life. Colin shares why the biggest issue was often not the rare disaster people remember. It was user-cancel, routine breakage, and the challenge of building long-term engagement.
Brian and Colin also discuss why humanoid robots may be solving the wrong problem for the home, why privacy has to be built in from the start, and why embodied AI can create a stronger connection than software alone.
If you want to understand what changed in robotics, and why the next wave of home robots may look very different from the last, this is the conversation.
KEY MOMENTS
(00:00) AI made impossible robots practical
(03:03) Why Colin did not retire
(03:37) The real bottleneck in robotics: situational awareness
(05:37) How reinforcement learning changed robot motion
(08:38) What Roomba taught Colin about routine
(09:38) Why people stopped using Roomba
(10:11) Why the poop problem was easier to solve
(11:34) Why embodied robots matter more than screens
(14:29) The MIT robot that helped inspire Aibo
(19:07) How you teach a robot social norms
(20:58) Training robots with stories
(22:44) Why consumer physical AI is the bigger opportunity
(25:55) Why a humanoid vacuum makes no sense
(28:21) What the first Familiar is designed to do
(31:31) How Colin thinks about value in the home
(36:18) Why cheaper hardware changed the equation
(38:14) When edge AI finally became possible
(40:15) Why home robots must run on the edge
(41:39) Why Roomba never got a real personality
(44:14) Why no two familiars will be the same
(48:10) What makes this robot different from past home robots
Connect with Colin Angle
https://www.linkedin.com/in/colinangle/
Learn more about Familiar Machines & Magic
https://www.familiarmachines.com/
Thoughts? Guest suggestions? Email us at podcast@automate.org
Find the transcript and more episodes at automated.fm
Unlock full access to Automated and explore everything automation.
Subscribe today on YouTube, Apple Podcasts, and Spotify.
So why did so many of them fail?
In this episode of Automated, Colin Angle explains why earlier generations of home robotics hit a wall, and why AI may finally change that.
After co-founding iRobot and helping bring Roomba into millions of homes, Colin is back with Familiar Machines & Magic, a new company focused on what he calls consumer physical AI.
He explains why the hardest problem in robotics was never just movement. It was perception, context, and the ability to understand what is happening in the real world.
The conversation also explores one of the biggest lessons from Roomba: even a technically impressive robot can fail if it does not fit naturally into daily life. Colin shares why the biggest issue was often not the rare disaster people remember. It was user-cancel, routine breakage, and the challenge of building long-term engagement.
Brian and Colin also discuss why humanoid robots may be solving the wrong problem for the home, why privacy has to be built in from the start, and why embodied AI can create a stronger connection than software alone.
If you want to understand what changed in robotics, and why the next wave of home robots may look very different from the last, this is the conversation.
KEY MOMENTS
(00:00) AI made impossible robots practical
(03:03) Why Colin did not retire
(03:37) The real bottleneck in robotics: situational awareness
(05:37) How reinforcement learning changed robot motion
(08:38) What Roomba taught Colin about routine
(09:38) Why people stopped using Roomba
(10:11) Why the poop problem was easier to solve
(11:34) Why embodied robots matter more than screens
(14:29) The MIT robot that helped inspire Aibo
(19:07) How you teach a robot social norms
(20:58) Training robots with stories
(22:44) Why consumer physical AI is the bigger opportunity
(25:55) Why a humanoid vacuum makes no sense
(28:21) What the first Familiar is designed to do
(31:31) How Colin thinks about value in the home
(36:18) Why cheaper hardware changed the equation
(38:14) When edge AI finally became possible
(40:15) Why home robots must run on the edge
(41:39) Why Roomba never got a real personality
(44:14) Why no two familiars will be the same
(48:10) What makes this robot different from past home robots
Connect with Colin Angle
https://www.linkedin.com/in/colinangle/
Learn more about Familiar Machines & Magic
https://www.familiarmachines.com/
Thoughts? Guest suggestions? Email us at podcast@automate.org
Find the transcript and more episodes at automated.fm
Unlock full access to Automated and explore everything automation.
Subscribe today on YouTube, Apple Podcasts, and Spotify.
Playback is via YouTube's official embedded player. Data from YouTube; Exumo is not affiliated with YouTube.