Loading video...

Video Failed to Load

Go Home

“Pretraining as we know it will end” Robotics: Hold my 🍺 , new data coming Source: Agibot

74,904 views • 1 year ago •via X (Twitter)

11 Comments

Appy Pie's profile picture
Appy Pie1 year ago

Ha! Robotics stepping in with a bang. The evolution of AI and robotics is bound to change the game in ways we can't fully predict. Exciting times ahead!

Fast Company's profile picture
Fast Company2 years ago

#AI agents: from tools to teammates. Software that can act, react, and interact with both humans and other agents will transform how businesses operate—and how they should invest in technology. This @EY_US article explains more. #ad

T's profile picture
T1 year ago

hello we would like to make… hello…?

sagi's profile picture
sagi1 year ago

Isn’t teleoperation technically post training data?

Yuxi on the Wired's profile picture
Yuxi on the Wired1 year ago

English website: Chinese website:

engineeringXroads's profile picture
engineeringXroads1 year ago

True

Eriiiiiiiick's profile picture
Eriiiiiiiick1 year ago

@xuxin_cheng your AI art is straight up next level, fam. What's the secret sauce behind your creative grind, and how do you think AI's gonna remix our whole vibe in the future?

Justin a AuDHDer… 🇨🇦's profile picture
Justin a AuDHDer… 🇨🇦1 year ago

What is the data and who come the video has no sound?

This is Jarvi's profile picture
This is Jarvi1 year ago

The data streams are about to 1000x

crypto_collector(🧀,🧀)'s profile picture
crypto_collector(🧀,🧀)1 year ago

So they just collecting real world data and how do they store it? All these different type of robots?

The New Money's profile picture
The New Money1 year ago

So, is it time to look a bit closer at robot stocks?

Related Videos

🔥 JUST IN: Open-source robotics dataset from 100% real-world scenarios! 🤯 Chinese robotics company AGIBOT just released AGIBOT WORLD 2026, an open-source dataset systematically covering key embodied AI research directions. Built entirely from real-world environments: commercial spaces, and homes. Collected using AGIBOT G2 robots in free-form collection mode, providing structured, accurately annotated, high-quality data. Digital twin technology creates 1:1 scale replicas in simulation matching the real environments. Both real-world and simulation data are open-sourced. The AGIBOT G2 platform collects multiple data types simultaneously: RGB(D) cameras, tactile sensors, force sensors, LiDAR, IMU, and full-body joint states. Whole-body control coordinates arms, waist, and hands for complex tasks. First-person teleoperation lets operators control the robot from its perspective. The tasks covered are fine-grained manipulation, ultra-long-horizon tasks, spatial navigation, dual-arm coordination, and multi-agent/human-robot collaboration. The dataset includes error-recovery trajectories with annotations. Most datasets only show successful demonstrations. AGIBOT includes failures and how the robot recovers, teaching models how to handle mistakes. After collection, data is tested through policy training and real-robot deployment to ensure quality. Then processed through industrial quality control with multiple screening and cleaning rounds. Making it open-source accelerates embodied AI research by giving researchers access to high-quality real-world robot data at scale. 🇨🇳 Learn more here: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

40,583 views • 4 months ago