Sergey Levine: Humanoid Robotics Results, Chinese Labs & Future Timelines
Ryan Peterman
59,569 views
Sergey Levine is one of the world's top robotics researchers and co-founder of Physical Intelligence. We talked about where humanoid robotics is today, thoughts on the Chinese robotics ecosystem, and his predictions for future timelines.
โข My ergonomic keyboard project I mentioned, you can follow along here: https://read.compose.llc/
โข The Kickstarter page for it: https://www.kickstarter.com/projects/...
๐ฃ๐ผ๐ฑ๐ฐ๐ฎ๐๐ ๐น๐ถ๐ป๐ธ๐:
โข Spotify: https://open.spotify.com/episode/19am...
โข Apple: https://podcasts.apple.com/us/podcast...
โข Transcript: https://www.developing.dev/p/sergey-l...
๐ง๐ต๐ฎ๐ป๐ธ ๐๐ผ๐ ๐๐ผ ๐๐ต๐ถ๐ ๐ฒ๐ฝ๐ถ๐๐ผ๐ฑ๐ฒ'๐ ๐๐ฝ๐ผ๐ป๐๐ผ๐ฟ ๐ณ๐ผ๐ฟ ๐๐๐ฝ๐ฝ๐ผ๐ฟ๐๐ถ๐ป๐ด ๐บ๐ ๐๐ผ๐ฟ๐ธ:
โข WorkOS: makes your app Enterprise Ready with easy to use APIs to add SSO, SCIM, RBAC, and more in just a few lines of code, check them out at https://workos.com/
๐ง๐ถ๐บ๐ฒ๐๐๐ฎ๐บ๐ฝ๐:
00:00 Intro
00:37 Where are we today
04:20 Most surprising capabilities so far
07:03 The most inspiring real world robotics
08:36 If OpenAI or Anthropic got into robotics
10:22 Chinese robotics
13:15 Will one lab breakout from the rest
16:59 Thoughts on a concrete roadmap
21:03 Generalization and demonstrating it
26:04 Types of data and which is best for robotics
34:34 Why humanoid robotics differs from Waymo
37:10 If humanoid robotics failed here is why
39:55 Are there hot take modeling architectures in robotics
42:05 Thoughts on AI safety in robotics
46:44 Top robotics research paper recommendation
49:35 Why is Boston Dynamics less top of mind
53:47 Advice for his younger self
56:42 Outro
๐ช๐ต๐ฒ๐ฟ๐ฒ ๐๐ผ ๐ณ๐ถ๐ป๐ฑ ๐ฆ๐ฒ๐ฟ๐ด๐ฒ๐:
โข Google Scholar: https://scholar.google.com/citations?...
โข Website: https://people.eecs.berkeley.edu/~svl...
โข Wikipedia: https://en.wikipedia.org/wiki/Sergey_...
โข X/Twitter: https://x.com/svlevine?lang=en
โข LinkedIn: LinkedIn: sergey-levine-5a31a24
๐ช๐ต๐ฒ๐ฟ๐ฒ ๐๐ผ ๐ณ๐ถ๐ป๐ฑ ๐ฅ๐๐ฎ๐ป:
โข Newsletter: https://www.developing.dev/
โข X/Twitter: https://x.com/ryanlpeterman
โข LinkedIn: LinkedIn: ryanlpeterman
โข Threads: https://www.threads.com/@ryanlpeterman
โข Instagram: Instagram: ryanlpeterman
โข TikTok: TikTok: ryanlpeterman
๐ฅ๐ฒ๐ณ๐ฒ๐ฟ๐ฒ๐ป๐ฐ๐ฒ๐ฑ ๐ถ๐ป ๐๐ต๐ถ๐ ๐ฒ๐ฝ๐ถ๐๐ผ๐ฑ๐ฒ:
โข Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware (ALOHA / ACT paper): https://arxiv.org/abs/2304.13705
โข Emergence of Human to Robot Transfer in Vision-Language-Action Models: https://arxiv.org/abs/2512.22414
โข Summary of human to robot paper: https://www.pi.website/research/human...
20 Comments
And thank you for what you do, please keep it up! Your videos are very much appreciated by junior engineers like me.
Would be great to hear from more people on the HW side. Model robustness is not there, but neither is mechanical - and for all use cases the practicality of HW reliability, service intervals, crash robustness etc are still IMO unsolved. And, dexterity / repeatable accuracy / load capability will always limit use in many cases, just because physics. So marketers will need to figure out the user needs based on those limitations before any vertical crosses the chasm.
I like the analogy about computers - we might see small elements of physical AI in many different places. Wallace and Gromit style. Ofc this is already the case in industrial automation.
]nterviews
I hope you can prepare deeper for interviews and ask the right questions in which the current robotics paradigm is trending towards, which is in-context learning for dexterous manipulation. Your interview questions and understanding led to an overall more shallow conversation that didnt extract as much deep insight from levineโs wealth of knowledge
Sergey Levine: Humanoid Robotics Results, Chinese Labs & Future Timelines
Ryan Peterman
59,569 views
Sergey Levine is one of the world's top robotics researchers and co-founder of Physical Intelligence. We talked about where humanoid robotics is today, thoughts on the Chinese robotics ecosystem, and his predictions for future timelines.
โข My ergonomic keyboard project I mentioned, you can follow along here: https://read.compose.llc/
โข The Kickstarter page for it: https://www.kickstarter.com/projects/...
๐ฃ๐ผ๐ฑ๐ฐ๐ฎ๐๐ ๐น๐ถ๐ป๐ธ๐:
โข Spotify: https://open.spotify.com/episode/19am...
โข Apple: https://podcasts.apple.com/us/podcast...
โข Transcript: https://www.developing.dev/p/sergey-l...
๐ง๐ต๐ฎ๐ป๐ธ ๐๐ผ๐ ๐๐ผ ๐๐ต๐ถ๐ ๐ฒ๐ฝ๐ถ๐๐ผ๐ฑ๐ฒ'๐ ๐๐ฝ๐ผ๐ป๐๐ผ๐ฟ ๐ณ๐ผ๐ฟ ๐๐๐ฝ๐ฝ๐ผ๐ฟ๐๐ถ๐ป๐ด ๐บ๐ ๐๐ผ๐ฟ๐ธ:
โข WorkOS: makes your app Enterprise Ready with easy to use APIs to add SSO, SCIM, RBAC, and more in just a few lines of code, check them out at https://workos.com/
๐ง๐ถ๐บ๐ฒ๐๐๐ฎ๐บ๐ฝ๐:
00:00 Intro
00:37 Where are we today
04:20 Most surprising capabilities so far
07:03 The most inspiring real world robotics
08:36 If OpenAI or Anthropic got into robotics
10:22 Chinese robotics
13:15 Will one lab breakout from the rest
16:59 Thoughts on a concrete roadmap
21:03 Generalization and demonstrating it
26:04 Types of data and which is best for robotics
34:34 Why humanoid robotics differs from Waymo
37:10 If humanoid robotics failed here is why
39:55 Are there hot take modeling architectures in robotics
42:05 Thoughts on AI safety in robotics
46:44 Top robotics research paper recommendation
49:35 Why is Boston Dynamics less top of mind
53:47 Advice for his younger self
56:42 Outro
๐ช๐ต๐ฒ๐ฟ๐ฒ ๐๐ผ ๐ณ๐ถ๐ป๐ฑ ๐ฆ๐ฒ๐ฟ๐ด๐ฒ๐:
โข Google Scholar: https://scholar.google.com/citations?...
โข Website: https://people.eecs.berkeley.edu/~svl...
โข Wikipedia: https://en.wikipedia.org/wiki/Sergey_...
โข X/Twitter: https://x.com/svlevine?lang=en
โข LinkedIn: LinkedIn: sergey-levine-5a31a24
๐ช๐ต๐ฒ๐ฟ๐ฒ ๐๐ผ ๐ณ๐ถ๐ป๐ฑ ๐ฅ๐๐ฎ๐ป:
โข Newsletter: https://www.developing.dev/
โข X/Twitter: https://x.com/ryanlpeterman
โข LinkedIn: LinkedIn: ryanlpeterman
โข Threads: https://www.threads.com/@ryanlpeterman
โข Instagram: Instagram: ryanlpeterman
โข TikTok: TikTok: ryanlpeterman
๐ฅ๐ฒ๐ณ๐ฒ๐ฟ๐ฒ๐ป๐ฐ๐ฒ๐ฑ ๐ถ๐ป ๐๐ต๐ถ๐ ๐ฒ๐ฝ๐ถ๐๐ผ๐ฑ๐ฒ:
โข Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware (ALOHA / ACT paper): https://arxiv.org/abs/2304.13705
โข Emergence of Human to Robot Transfer in Vision-Language-Action Models: https://arxiv.org/abs/2512.22414
โข Summary of human to robot paper: https://www.pi.website/research/human...