How can AI systems understand the physical world the way humans do? 🌍 On Feb. 5 at 9 a.m. PT, join NVIDIA and IntBot engineers for a live session exploring how to fine‑tune reasoning for complex, dynamic settings like heavy traffic, filter synthetic data for reliability, and

Noticeable improvements in Grok. Also enjoying X video.

Hot take from investing in robotics companies: simulation is overrated. Billions poured into sim-to-real. The gap hasn't meaningfully closed in 2-3 years. Locomotion tolerates 10% failure rates — manipulation doesn't. The industry is quietly converging on: bulk imitation data

A walk in the forest increases immunity for 30 days. Trees emit phytoncides. By inhaling them for 2 hours, you increase the number and activity of NK cells (cancer and virus killers) in the blood by 50%. The effect lasts for a month. This is the Japanese practice of

🚨 ALIBABA JUST OPEN-SOURCED THE MOST POWERFUL MODEL IN THE WORLD - Qwen3.5 plus And it changes the AI race forever. This is not another “bigger model” release. This is a complete architectural reset. 397B parameters → only 17B active at inference, Qwen3.5 plus delivers

Major cheat code for life: Increase your recovery speed. You will get rejected. You will lose money. You will embarrass yourself. The goal isn't to avoid the fall. It's to shorten the time between the fall and the reset. Fast recovery compounds.

Today's state-of-the-art VLAs struggle to generalize zero-shot to new robot embodiments, despite training on extensive multi-embodiment data. We introduce Language-Action Pre-training (LAP) and LAP-3B — the first VLA to achieve substantial zero-shot transfer to unseen real-world

🤖 One predictive backbone, three distinct tasks, consistent gains: a strong signal that investing in reusable world models is the right abstraction for robot learning. Toyota Research Institute (TRI) just ran the same idea three different ways—and it worked each time. Using our NVIDIA Cosmos

robots building robots… •unitree dropped a video of their g1 humanoid assembling robot legs in their own shenzhen factory using the unifolm embodied ai model. yes, robots building robots at actual production speed •the tasks are real bimanual manipulation, picking parts and

X-Humanoid (Beijing Innovation Center of Humanoid Robotics) has officially launched Embodied Tien Kung 3.0, a general-purpose humanoid platform designed to lower technical barriers for developers and researchers. - The company claims it is the first full-size humanoid to

You can buy a humanoid robot on amazon now

Announcing DreamDojo: our open-source, interactive world model that takes robot motor controls and generates the future in pixels. No engine, no meshes, no hand-authored dynamics. It's Simulation 2.0. Time for robotics to take the bitter lesson pill. Real-world robot learning is

We are still so early. Literally.

In the current state of VLAs, how do ppl handle recovery and task progress? I don’t see any VLA models using any form of explicit environment state (other than observations) as input, which could indicate how much of the task has been completed. Recovery is demonstrated in the

JUST IN 🔴 A new U.S. military led task force focused on cartel intelligence played a role in Mexico’s operation targeting cartel boss “El Mencho,” a U.S. official said. According to CBS News, a Pentagon joint task force that routinely coordinates with the Mexican military took

ajc.com/business/2026/…

Army kills 'El Tuli', El Mencho's right-hand man who masterminded the post-boss-death violence, blockades & fires in Jalisco. He even put a 20k peso (~$1,200) bounty on soldiers' heads. Killed in shootout while fleeing; troops seized ~$410K MXN + $965K USD cash + arsenal. Big

Calling all researchers! 🤖The CoRL 2026 website is officially live at corl.org with key dates for your submissions: 🗓 May 25: Abstract Submission 🗓 May 28: Full Paper Submission 🗓 Nov 9-12: Conference in Austin, TX Send us your coolest work! #RobotLearning