Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

🚨 Many cloud-dependent ROBOT systems stop working if the internet is disconnected Because they depend on: → Cloud APIs → Remote servers → Constant connectivity But this ROBOT dog didn't stop working... 🤯 And the camera data can be processed LOCALLY too 🔒 This changes everything for real-world robotics.... show more

MAPUNDA

42,021 subscribers

157,362 views • 1 month ago •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

This isn’t a coffee-fetching robot — it’s a life-saving robot. Developed at Carnegie Mellon University in Pittsburgh, this robot dog is designed to help medics in emergencies too dangerous for humans. It can see in the dark, pick up sounds and even be pet. But even its creators admit: it’s intimidating to many humans— and they’re still working on making it feel more approachable. Tony Dokoupil got a firsthand look.

This isn’t a coffee-fetching robot — it’s a life-saving robot. Developed at Carnegie Mellon University in Pittsburgh, this robot dog is designed to help medics in emergencies too dangerous for humans. It can see in the dark, pick up sounds and even be pet. But even its creators admit: it’s intimidating to many humans— and they’re still working on making it feel more approachable. Tony Dokoupil got a firsthand look.

CBS Evening News with Tony Dokoupil

14,493 views • 7 months ago

Zerith Robotics in China has started mass production of its H1 humanoid robot for real-world use. It is designed for places like hotels and stores, moving on wheels and working on its own indoors. The robot can clean spaces and also help in supermarkets by picking and handling products automatically.

Zerith Robotics in China has started mass production of its H1 humanoid robot for real-world use. It is designed for places like hotels and stores, moving on wheels and working on its own indoors. The robot can clean spaces and also help in supermarkets by picking and handling products automatically.

Space and Technology

14,656 views • 3 months ago

Guess what I can finally talk about the Sail Forth RELEASE DATE and it is 🚨🚨 RIGHT NOW!!! 🚨🚨 I've been working on this game for over 6 years, this doesn't even seem real lol. It would mean the world to me if you could share this! Link and thread about the game below 👇

Guess what I can finally talk about the Sail Forth RELEASE DATE and it is 🚨🚨 RIGHT NOW!!! 🚨🚨 I've been working on this game for over 6 years, this doesn't even seem real lol. It would mean the world to me if you could share this! Link and thread about the game below 👇

festive vector

246,269 views • 3 years ago

Watch this robot dog learn to walk from scratch in real time! Our new method, APRL, dynamically adjusts exploration constraints to enable fast and performant RL directly in the real world. APRL can also adapt to changes in the terrain. No simulation, no demos. A thread 👇

Watch this robot dog learn to walk from scratch in real time! Our new method, APRL, dynamically adjusts exploration constraints to enable fast and performant RL directly in the real world. APRL can also adapt to changes in the terrain. No simulation, no demos. A thread 👇

Sergey Levine

105,579 views • 2 years ago

This robot assistant from the NVIDIA CES Keynote on Monday is going viral. Nader Khalil🍊 explains all the hottest emerging AI trends in one demo: AI applications in 2026 will be multi-model, multi-modal, hybrid cloud/local, use open source models as well as proprietary models, control robots and embedded devices in the physical world, and have voice interfaces. (And the demo had a cute robot *and* a cute dog. Gold.) The demo was built with Pipecat AI. NVIDIA posted a really nice technical walk-through and complete code. The Reachy Mini robot from Hugging Face is open source hardware. (You can order it now, I have one!). You can run the assistant locally on your own hardware, in the cloud, or both.

This robot assistant from the NVIDIA CES Keynote on Monday is going viral. Nader Khalil🍊 explains all the hottest emerging AI trends in one demo: AI applications in 2026 will be multi-model, multi-modal, hybrid cloud/local, use open source models as well as proprietary models, control robots and embedded devices in the physical world, and have voice interfaces. (And the demo had a cute robot and a cute dog. Gold.) The demo was built with Pipecat AI. NVIDIA posted a really nice technical walk-through and complete code. The Reachy Mini robot from Hugging Face is open source hardware. (You can order it now, I have one!). You can run the assistant locally on your own hardware, in the cloud, or both.

kwindla

49,028 views • 7 months ago

I asked the server at this sushi place if the robot servers stole her shifts and her answer surprised me. a few years ago they brought in robot servers. The boss started with 1 robot, cut the staff from 5 to 4. From day 1 the robots were a mess. They can basically only deliver drinks. They get in the way. The 4 remaining servers felt about 20% busier, like the robot made no difference. But the kids LOVED them. Adults posted videos to social media like I'm doing right now. This grew their business in an unexpected way. Today they only have 2 robots. Zero shifts replaced (she says). But the robots still deliver drinks and the restaurant is busier than ever. Conveyor belt sushi is brilliant. Familiar to Japan for decades, newer to the US. The more plates you grab, the more prizes you win. Gamification in real life. What this restaurant did is genius. Talk about a flywheel: - Conveyor belt drives social media virality, increasing sales - Robot servers drive social media virality, increasing sales - Gamification drives average order values (The food is pretty garbage BTW, but my kids love the concept and it's a tradition for us) The server was hesitant to admit this, but after working out the robot kinks it sounded like there would be more human shifts today if not for the 2 robots. How many people order drinks JUST to see the robot? They didn't replace humans on a 1:1 ratio. But 1:3 or 1:4, yes. AI and automation will add much more in efficiency and productivity than the jobs it takes. A huge net win.

I asked the server at this sushi place if the robot servers stole her shifts and her answer surprised me. a few years ago they brought in robot servers. The boss started with 1 robot, cut the staff from 5 to 4. From day 1 the robots were a mess. They can basically only deliver drinks. They get in the way. The 4 remaining servers felt about 20% busier, like the robot made no difference. But the kids LOVED them. Adults posted videos to social media like I'm doing right now. This grew their business in an unexpected way. Today they only have 2 robots. Zero shifts replaced (she says). But the robots still deliver drinks and the restaurant is busier than ever. Conveyor belt sushi is brilliant. Familiar to Japan for decades, newer to the US. The more plates you grab, the more prizes you win. Gamification in real life. What this restaurant did is genius. Talk about a flywheel: - Conveyor belt drives social media virality, increasing sales - Robot servers drive social media virality, increasing sales - Gamification drives average order values (The food is pretty garbage BTW, but my kids love the concept and it's a tradition for us) The server was hesitant to admit this, but after working out the robot kinks it sounded like there would be more human shifts today if not for the 2 robots. How many people order drinks JUST to see the robot? They didn't replace humans on a 1:1 ratio. But 1:3 or 1:4, yes. AI and automation will add much more in efficiency and productivity than the jobs it takes. A huge net win.

Chris Koerner

69,205 views • 12 days ago

Today we are excited to open up Neuracore to the academic community! Neuracore is a new data foundation built to accelerate robot learning by removing one of the field’s biggest bottlenecks: capturing and working with high-fidelity multimodal robotics data. For the first time, researchers can store, view, and work with robotics data in a cloud-native system built specifically for large-scale learning, and we are making this core platform completely free for academia. The platform lets teams capture every sensor at its native rate, store and visualize data without loss, and then train and deploy models locally using our open-source code (Link in the comments). We are rolling out access to select academic institutions first. Anyone with an academic email can sign up, and if your institution is not part of the initial rollout, you will be able to join the waitlist directly. Beyond providing this infrastructure, we see an opportunity to build a global community where engineers and researchers can share, collaborate, and advance the frontier of robot learning together. Supported by our recent $3M pre-seed round led by Earlybird VC, we are excited to take this mission even further. Our long-term goal is for Neuracore to become the natural home for cutting-edge robot learning algorithms and real-world robotics experimentation, helping accelerate the next wave of Physical AI.

Today we are excited to open up Neuracore to the academic community! Neuracore is a new data foundation built to accelerate robot learning by removing one of the field’s biggest bottlenecks: capturing and working with high-fidelity multimodal robotics data. For the first time, researchers can store, view, and work with robotics data in a cloud-native system built specifically for large-scale learning, and we are making this core platform completely free for academia. The platform lets teams capture every sensor at its native rate, store and visualize data without loss, and then train and deploy models locally using our open-source code (Link in the comments). We are rolling out access to select academic institutions first. Anyone with an academic email can sign up, and if your institution is not part of the initial rollout, you will be able to join the waitlist directly. Beyond providing this infrastructure, we see an opportunity to build a global community where engineers and researchers can share, collaborate, and advance the frontier of robot learning together. Supported by our recent $3M pre-seed round led by Earlybird VC, we are excited to take this mission even further. Our long-term goal is for Neuracore to become the natural home for cutting-edge robot learning algorithms and real-world robotics experimentation, helping accelerate the next wave of Physical AI.

Neuracore

40,796 views • 8 months ago

$EITHER | Eitherway There’s so many reasons to be super BULLISH on $EITHER. This Solana utility is showcasing its real world application and so much more. Believing 100mil in the next stop on the way to becoming a blue chip project very soon. This team is legit and non stop working on Eitherway

$EITHER | Eitherway There’s so many reasons to be super BULLISH on $EITHER. This Solana utility is showcasing its real world application and so much more. Believing 100mil in the next stop on the way to becoming a blue chip project very soon. This team is legit and non stop working on Eitherway

Drew

26,009 views • 3 months ago

Feeling what a robot feels is becoming real! 🫰 Fluid Reality just ran its first full end-to-end touch teleoperation system, and the demo shows why high-resolution haptics might be the missing piece in robot control. A 22 mm fingertip display with 32 independently actuated “bubbles” was mapped directly to tactile sensors on a robot hand. 🫧 Three fingers streamed real-time touch data robot → human, giving the operator actual contact feedback instead of waving their hands blind, like most teleop systems today. It’s still early, but this is a real step toward dexterous ops and high-quality data for training Physical AI systems. Companies involved: Samsung Electronics, Alt-Bionics, Inc. (the robot hand), MANUS™ (hand tracking), and Sensible Robotics (fingertip sensors) Touch isn’t a nice-to-have in robotics. It’s the difference between manipulation that works in the real world P.S. Guess how many touch receptors we humans, have in our hand! :) ~~ ♻ Join the weekly robotics newsletter, and never miss any news →

Feeling what a robot feels is becoming real! 🫰 Fluid Reality just ran its first full end-to-end touch teleoperation system, and the demo shows why high-resolution haptics might be the missing piece in robot control. A 22 mm fingertip display with 32 independently actuated “bubbles” was mapped directly to tactile sensors on a robot hand. 🫧 Three fingers streamed real-time touch data robot → human, giving the operator actual contact feedback instead of waving their hands blind, like most teleop systems today. It’s still early, but this is a real step toward dexterous ops and high-quality data for training Physical AI systems. Companies involved: Samsung Electronics, Alt-Bionics, Inc. (the robot hand), MANUS™ (hand tracking), and Sensible Robotics (fingertip sensors) Touch isn’t a nice-to-have in robotics. It’s the difference between manipulation that works in the real world P.S. Guess how many touch receptors we humans, have in our hand! :) ~~ ♻ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

46,927 views • 8 months ago

NAVIAI-I3, a more robust humanoid robot, was launched by the Zhejiang Humanoid Robotics Innovation Center. Standing 1.8 meters tall, it can run like a human and walk on various terrains. This humanoid robot appears to be primarily designed for specialized industrial tasks, such as security patrols in underground parking lots. They previously launched the lightweight bipedal NVAI-2 and a wheeled humanoid robot. The bipedal robot is used in manufacturing operations, while the wheeled humanoid robot is used for unstructured, flexible tasks such as warehouse sorting, supermarket shelving, and office cleaning. We filmed this demonstration at the World Robot Conference; you can see it in the comments section.

NAVIAI-I3, a more robust humanoid robot, was launched by the Zhejiang Humanoid Robotics Innovation Center. Standing 1.8 meters tall, it can run like a human and walk on various terrains. This humanoid robot appears to be primarily designed for specialized industrial tasks, such as security patrols in underground parking lots. They previously launched the lightweight bipedal NVAI-2 and a wheeled humanoid robot. The bipedal robot is used in manufacturing operations, while the wheeled humanoid robot is used for unstructured, flexible tasks such as warehouse sorting, supermarket shelving, and office cleaning. We filmed this demonstration at the World Robot Conference; you can see it in the comments section.

CyberRobo

15,521 views • 9 months ago

ChatGPT reached 365 billion searches 5.5 times faster than Google. a16z GP David George: “AI is built on the back of the internet and cloud computing. And because of that, it allows for immediate global distribution.” “Because of the nature of this technology, you don't have to deliver a new hardware product. And because we have global internet buildout, and because we have cloud computing, the whole world can access this.” David George

ChatGPT reached 365 billion searches 5.5 times faster than Google. a16z GP David George: “AI is built on the back of the internet and cloud computing. And because of that, it allows for immediate global distribution.” “Because of the nature of this technology, you don't have to deliver a new hardware product. And because we have global internet buildout, and because we have cloud computing, the whole world can access this.” David George

a16z

60,448 views • 6 months ago

Happy to share what I’ve been working on since joining Genesis! GENE-26.5 is a one-of-a-kind, robotics-native multimodal foundation model that learns from diverse, in-the-wild data across modalities and outputs actions enabling a 54-DoF robot system to perform the most dexterous, long-horizon manipulation tasks to date—approaching human-level capability. This is the result of innovations across the full stack—data collection and processing, robot systems, model architecture, training strategies, and scalable evaluation infrastructure.

Happy to share what I’ve been working on since joining Genesis! GENE-26.5 is a one-of-a-kind, robotics-native multimodal foundation model that learns from diverse, in-the-wild data across modalities and outputs actions enabling a 54-DoF robot system to perform the most dexterous, long-horizon manipulation tasks to date—approaching human-level capability. This is the result of innovations across the full stack—data collection and processing, robot systems, model architecture, training strategies, and scalable evaluation infrastructure.

Zu Wang

19,673 views • 3 months ago

Introducing WorldForge: testable world-model workflows for physical AI systems. You can think of it, loosely, as “LangChain for world models”. The problem is that “world model” has become an overloaded label. Depending on context, it can mean a video generator, a cost model, a robot policy, a JEPA-style latent predictor, etc. They share almost nothing, different inputs, runtimes, failure modes. I built WorldForge to stop pretending they're interchangeable. Front-door demo: a real Hugging Face LeRobot (LeRobot) diffusion_pusht policy combined with LeWorldModel by Lucas Maes checkpoint for scoring. Both run locally on my MacBook in the demo video. LeWM is extremely efficient (~15M params), can plan up to 48× faster, and runs on commodity hardware. WorldForge wires the loop: policy → candidates → score → select Replay happens in a local TUI today, but the same loop could drive a real robot in the world. Would love feedback from people working on world models, physical AI, robotics, ML infra, and adjacent tooling. Fully open source. Contributions very welcome. Plan in the dream, replay in real world.

Introducing WorldForge: testable world-model workflows for physical AI systems. You can think of it, loosely, as “LangChain for world models”. The problem is that “world model” has become an overloaded label. Depending on context, it can mean a video generator, a cost model, a robot policy, a JEPA-style latent predictor, etc. They share almost nothing, different inputs, runtimes, failure modes. I built WorldForge to stop pretending they're interchangeable. Front-door demo: a real Hugging Face LeRobot (LeRobot) diffusion_pusht policy combined with LeWorldModel by Lucas Maes checkpoint for scoring. Both run locally on my MacBook in the demo video. LeWM is extremely efficient (~15M params), can plan up to 48× faster, and runs on commodity hardware. WorldForge wires the loop: policy → candidates → score → select Replay happens in a local TUI today, but the same loop could drive a real robot in the world. Would love feedback from people working on world models, physical AI, robotics, ML infra, and adjacent tooling. Fully open source. Contributions very welcome. Plan in the dream, replay in real world.

abdel

17,001 views • 3 months ago

Have you ever heard of the US CLOUD Act? No? It means the US government can demand data from American cloud companies, even when the servers are in another country. And Oracle is already embedded across the UK’s government plumbing: NHS systems, tax records, benefits, employment, immigration, justice, and defense cloud. So when King Charles confirmed today that digital ID is still moving forward, just remember: this is not only a UK surveillance story. It is also an American one. Didn’t we fight a whole war to stop being one country? So why is our data infrastructure doing Pangea cosplay?

Have you ever heard of the US CLOUD Act? No? It means the US government can demand data from American cloud companies, even when the servers are in another country. And Oracle is already embedded across the UK’s government plumbing: NHS systems, tax records, benefits, employment, immigration, justice, and defense cloud. So when King Charles confirmed today that digital ID is still moving forward, just remember: this is not only a UK surveillance story. It is also an American one. Didn’t we fight a whole war to stop being one country? So why is our data infrastructure doing Pangea cosplay?

Drey

13,869 views • 3 months ago

We might be solving the wrong problem in robotics. That’s what this makes clear. UMI → Universal Manipulation Interface A simple $400 gripper that lets you teach robots by demonstration. You hold it like a tool. Show the task. The robot learns. No teleoperation. No expensive hardware. No robot-specific data. Stanford open-sourced everything → hardware, code, datasets. What stands out to me is the bottleneck. Not algorithms. Data. Teleoperation → ~35 demos/hour UMI → ~111 demos/hour And the data transfers across robots → UR5, Franka, others. The design is surprisingly practical: → GoPro fisheye lens (155° FOV) + mirrors for depth → SLAM + IMU for precise 6DoF tracking → latency matching for dynamic tasks → diffusion policies for multimodal actions Then it scales. Cheng Chi takes this further with Sunday Robotics (with Tony Zhao). A $200 glove → deployed in 500+ homes → ~10 million real-world interactions. Not lab data. Real human behavior. Their robot learns dishes, laundry, espresso → with zero robot-specific data. This is where the shift becomes obvious. From training robots in controlled environments → to learning directly from humans at scale So here’s the real question: Will robotics be unlocked by better models… or by unlocking data? #ArtificialIntelligence #Robotics #AI #Innovation #FutureOfWork

We might be solving the wrong problem in robotics. That’s what this makes clear. UMI → Universal Manipulation Interface A simple $400 gripper that lets you teach robots by demonstration. You hold it like a tool. Show the task. The robot learns. No teleoperation. No expensive hardware. No robot-specific data. Stanford open-sourced everything → hardware, code, datasets. What stands out to me is the bottleneck. Not algorithms. Data. Teleoperation → ~35 demos/hour UMI → ~111 demos/hour And the data transfers across robots → UR5, Franka, others. The design is surprisingly practical: → GoPro fisheye lens (155° FOV) + mirrors for depth → SLAM + IMU for precise 6DoF tracking → latency matching for dynamic tasks → diffusion policies for multimodal actions Then it scales. Cheng Chi takes this further with Sunday Robotics (with Tony Zhao). A $200 glove → deployed in 500+ homes → ~10 million real-world interactions. Not lab data. Real human behavior. Their robot learns dishes, laundry, espresso → with zero robot-specific data. This is where the shift becomes obvious. From training robots in controlled environments → to learning directly from humans at scale So here’s the real question: Will robotics be unlocked by better models… or by unlocking data? #ArtificialIntelligence #Robotics #AI #Innovation #FutureOfWork

Pascal Bornet

186,551 views • 4 months ago

Here's the news- I'm moving on from my head of community role at the Data on Kubernetes Community. Too many people to thank and I'm extremely grateful for having been able to watch this community grow the last 2.5 years. I'll stay working in the cloud native ecosystem- let's stay in touch!

Here's the news- I'm moving on from my head of community role at the Data on Kubernetes Community. Too many people to thank and I'm extremely grateful for having been able to watch this community grow the last 2.5 years. I'll stay working in the cloud native ecosystem- let's stay in touch!

Bart Farrell

10,001 views • 3 years ago

Policies trained on real robot data via imitation can be surprisingly capable. But for domains like dexterous manipulation, they are often not quite good enough: they move slowly, miss grasps, make unreliable contact, and fail under small perturbations. Can we improve them without any additional data collection on the real robot? In SCORE, we show that we can improve real-world diffusion/flow policies cheaply by using simulation to simply learn how to steer them on deployment. This leads to large gains in real-world success and speed across a variety of tasks, without requiring additional real-world experience: 🧵 (1/10)

Policies trained on real robot data via imitation can be surprisingly capable. But for domains like dexterous manipulation, they are often not quite good enough: they move slowly, miss grasps, make unreliable contact, and fail under small perturbations. Can we improve them without any additional data collection on the real robot? In SCORE, we show that we can improve real-world diffusion/flow policies cheaply by using simulation to simply learn how to steer them on deployment. This leads to large gains in real-world success and speed across a variety of tasks, without requiring additional real-world experience: 🧵 (1/10)

Abhishek Gupta

34,775 views • 1 month ago

Okay, full 5 second laugh break done. But wait… This tiny DIY robot made in China is already learning how to grab, move, and manipulate toys autonomously inside a real home environment. That is literally the first real step toward personal robotics and autonomous home manipulation. Toys are officially entering the robot era. At first you laugh, then two seconds later you realize this is exactly how robotics culture spreads: from labs running fully autonomous humanoids to factories to garages and now directly into kids’ bedrooms with DIY robots already doing miniature warehouse robotics. This is physical AI leaving the cloud and entering the living room. And honestly, it’s exciting watching this happen live on X right now.

Okay, full 5 second laugh break done. But wait… This tiny DIY robot made in China is already learning how to grab, move, and manipulate toys autonomously inside a real home environment. That is literally the first real step toward personal robotics and autonomous home manipulation. Toys are officially entering the robot era. At first you laugh, then two seconds later you realize this is exactly how robotics culture spreads: from labs running fully autonomous humanoids to factories to garages and now directly into kids’ bedrooms with DIY robots already doing miniature warehouse robotics. This is physical AI leaving the cloud and entering the living room. And honestly, it’s exciting watching this happen live on X right now.

TechniaHQ | humanoid robots

18,738 views • 3 months ago

Train robots. Earn rewards. Base is pioneering the Train-to-Earn model for robotics. With Axis Robotics, anyone can help train robot simulations from a laptop and be rewarded for contributing data. The future of robotics won’t be trained by a handful of companies. It’ll be trained by everyone. PL 漂亮 (✱,✱) showcased this vision at SuperAI, the largest AI event in Asia 👇

Train robots. Earn rewards. Base is pioneering the Train-to-Earn model for robotics. With Axis Robotics, anyone can help train robot simulations from a laptop and be rewarded for contributing data. The future of robotics won’t be trained by a handful of companies. It’ll be trained by everyone. PL 漂亮 (✱,✱) showcased this vision at SuperAI, the largest AI event in Asia 👇

Base APAC

18,980 views • 2 months ago

Your phone can build voxel world now! 👾 This is an open-source tool that converts phone video into 3D point clouds. You can run Depth Anything 3 locally. Basically just local GPU inference. The interesting part is how it handles messy input. Video windows overlap, scale gets converted to real meters, drift gets corrected after inference. The system stitches everything together even when footage is noisy. Objects far away might fail strict confidence filters but still show up as recognizable buildings. That's huge for small robots because perception becomes cheap and stays local. Obviously it wouldn't be a phone for robots, but it shows the potential of edge/local perception processing. A robot can understand its environment without offloading to the cloud. Real-world video has artifacts and occlusions. The system doesn't need perfect input to reconstruct scene structure. It works with what's there. For robotics, this means small platforms can build 3D maps autonomously using hardware they already have. Here's the repo: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Your phone can build voxel world now! 👾 This is an open-source tool that converts phone video into 3D point clouds. You can run Depth Anything 3 locally. Basically just local GPU inference. The interesting part is how it handles messy input. Video windows overlap, scale gets converted to real meters, drift gets corrected after inference. The system stitches everything together even when footage is noisy. Objects far away might fail strict confidence filters but still show up as recognizable buildings. That's huge for small robots because perception becomes cheap and stays local. Obviously it wouldn't be a phone for robots, but it shows the potential of edge/local perception processing. A robot can understand its environment without offloading to the cloud. Real-world video has artifacts and occlusions. The system doesn't need perfect input to reconstruct scene structure. It works with what's there. For robotics, this means small platforms can build 3D maps autonomously using hardware they already have. Here's the repo: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

50,485 views • 1 month ago