Title: Scaling Robot Learning Beyond Teleoperation with EgoVerse...
URL Source: https://www.bestblogs.dev/status/2036136375494517142
Published Time: 2026-03-23 17:41:56
Markdown Content: Skip to main content Toggle navigation menu Toggle navigation menuArticlesPodcastsVideosTweetsSourcesNewsletters
⌘K
Change language Switch ThemeSign In
Narrow Mode
Scaling Robot Learning Beyond Teleoperation with EgoVerse
Scaling Robot Learning Beyond Teleoperation with EgoVerse
 ### Jim Fan@DrJimFan
Teleop is so 2025. Ever since we unveiled EgoScale and the dexterity scaling law, it's been clear to us and the ecosystem that behavior cloning directly from humans is the way to break the curse of teleop. 2026 is all about scaling robot learning without robots.
#### Danfei Xu
@danfei_xu · 4h ago
Introducing EgoVerse: an ecosystem for robot learning from egocentric human data.
Built and tested by 4 research labs + 3 industry partners, EgoVerse enables both science and scaling
1300+ hrs, 240 scenes, 2000+ tasks, and growing
Dataset design, findings, and ecosystem 🧵Show More
00:59
18
73
356
51.1K
Mar 23, 2026, 5:41 PM View on X
22 Replies
18 Retweets
170 Likes
20.5K Views  Jim Fan @DrJimFan
One Sentence Summary
Jim Fan discusses the industry shift from teleoperation to behavior cloning, leveraging the new EgoVerse ecosystem to scale robot learning without physical robots.
Summary
Jim Fan, NVIDIA's Director of Robotics, provides an expert perspective on the evolution of robot learning. He argues that the field is moving away from teleoperation toward behavior cloning, a shift accelerated by the introduction of EgoVerse—an ecosystem for learning from egocentric human data. This approach aligns with the 'dexterity scaling law,' signaling a strategic move toward scaling robot capabilities without relying on direct human teleoperation.
AI Score
83
Influence Score 37
Published At Today
Language
English
Tags
Robotics
RobotLearning
EgoVerse
BehaviorCloning
ScalingLaws HomeArticlesPodcastsVideosTweets