正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

💥Introducing FACTR 2, learning external force sensing on commodity robot arms without needing dedicated sensors. We show that learned force signals enable force-feedback teleop on low-cost arms and improve BC policies. FACTR 2 consists of: 1. Neural External Torque (NEXT): learns external forces without needing dedicated force sensors. 2.... show more

Jason Liu

2,110 subscribers

109,194 次观看 • 1 个月前 •via X (Twitter)

教育科学技术

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Low-cost teleop systems have democratized robot data collection, but they lack any force feedback, making it challenging to teleoperate contact-rich tasks. Many robot arms provide force information — a critical yet underutilized modality in robot learning. We introduce: 1. 🦾A low-cost, force-feedback-enabled teleop system. 2. 🥊Force-Attending Curriculum Training (FACTR) uses force to improve generalization in complex, contact-rich tasks. 🧵(1/N)

Low-cost teleop systems have democratized robot data collection, but they lack any force feedback, making it challenging to teleoperate contact-rich tasks. Many robot arms provide force information — a critical yet underutilized modality in robot learning. We introduce: 1. 🦾A low-cost, force-feedback-enabled teleop system. 2. 🥊Force-Attending Curriculum Training (FACTR) uses force to improve generalization in complex, contact-rich tasks. 🧵(1/N)

Jason Liu

151,921 次观看 • 1 年前

Force is arguably the most overlooked ingredient in modern robot learning. Introducing FACTR 2: it turns *any* commodity robot into a force-aware system with no force sensors required. Train a tiny force network in <1min with <10mins of data and drop it into any existing teleop pipelines: ✅ Free force sensing for both the robot and the operator arm ✅ Makes demos higher-quality → fewer of them needed. ✅ A new force-aware learning algorithm (FIRST) uses those recovered forces to figure out which parts of a demo actually matter, making learning data-efficient. ✅ Strong performance on complex tasks with fewer demos and even no pretraining! More details below.

Force is arguably the most overlooked ingredient in modern robot learning. Introducing FACTR 2: it turns any commodity robot into a force-aware system with no force sensors required. Train a tiny force network in <1min with <10mins of data and drop it into any existing teleop pipelines: ✅ Free force sensing for both the robot and the operator arm ✅ Makes demos higher-quality → fewer of them needed. ✅ A new force-aware learning algorithm (FIRST) uses those recovered forces to figure out which parts of a demo actually matter, making learning data-efficient. ✅ Strong performance on complex tasks with fewer demos and even no pretraining! More details below.

Deepak Pathak

38,418 次观看 • 1 个月前

Excited to share that our work NeuralActuator: Neural Actuation Modeling for Robot Dynamics and External Force Perception has been accepted to #RSS2026! Your robot — even a low-cost one — can feel external forces without torque or tactile sensors. TL;DR: NeuralActuator is a neural actuator model that jointly predicts 1️⃣torque to capture the nonlinear and time-varying current–to–torque relationship of low-cost servos, 2️⃣external contact forces (and force detection gates) for sensorless force perception, 3️⃣and motor conditions that indicate each motor’s operating regime. Here is a fast-forward video clip ⬇️ We are also covering more robots like LeRobot-S101 and Franka Panda. More details coming soon.

Excited to share that our work NeuralActuator: Neural Actuation Modeling for Robot Dynamics and External Force Perception has been accepted to #RSS2026! Your robot — even a low-cost one — can feel external forces without torque or tactile sensors. TL;DR: NeuralActuator is a neural actuator model that jointly predicts 1️⃣torque to capture the nonlinear and time-varying current–to–torque relationship of low-cost servos, 2️⃣external contact forces (and force detection gates) for sensorless force perception, 3️⃣and motor conditions that indicate each motor’s operating regime. Here is a fast-forward video clip ⬇️ We are also covering more robots like LeRobot-S101 and Franka Panda. More details coming soon.

Zhiyang (Frank) Dou

50,481 次观看 • 3 个月前

1/🧠Humans are the best robot data source — but video alone misses one thing: force. 2/🙁Tactile gloves capture force — but they're costly and block the real touch manipulation depends on. 3/💪Maybe the future of touch lives on your wrist: surface EMG reads the muscles that cause force — tactile sensing without ever touching a tactile sensor. 4/🔥Want a fully open-source framework — hardware + software — to train your own force-aware learn-from-human-data robot policy? 🚀We introduce ForceBand: Learning Forceful Manipulation with sEMG -- bring force into human videos with sEMG, for force-aware manipulation ⬇️ ✦ Zero-Shot Human-to-Robot Transfer ✦ Force Beyond Vision ✦ Free-Hand Force Sensing ✦ Collect by Anyone, Anytime, Anywhere ✦ Deploy on Any Robot, Any Camera, Any Environment ✦ Open-Source & Low-Cost & Easy-to-Implement Let's squeeze every bit of signal out of human data, and let robots feel the force! 🌐 Website: 📄 Paper: 💻 Code: 🎥 Video: 🧵 1/n

1/🧠Humans are the best robot data source — but video alone misses one thing: force. 2/🙁Tactile gloves capture force — but they're costly and block the real touch manipulation depends on. 3/💪Maybe the future of touch lives on your wrist: surface EMG reads the muscles that cause force — tactile sensing without ever touching a tactile sensor. 4/🔥Want a fully open-source framework — hardware + software — to train your own force-aware learn-from-human-data robot policy? 🚀We introduce ForceBand: Learning Forceful Manipulation with sEMG -- bring force into human videos with sEMG, for force-aware manipulation ⬇️ ✦ Zero-Shot Human-to-Robot Transfer ✦ Force Beyond Vision ✦ Free-Hand Force Sensing ✦ Collect by Anyone, Anytime, Anywhere ✦ Deploy on Any Robot, Any Camera, Any Environment ✦ Open-Source & Low-Cost & Easy-to-Implement Let's squeeze every bit of signal out of human data, and let robots feel the force! 🌐 Website: 📄 Paper: 💻 Code: 🎥 Video: 🧵 1/n

Zhi (Leo) Wang

50,356 次观看 • 1 个月前

What if some parts of a robot demonstration are more important than others? Most of a trajectory is free-space motion. But success or failure is often determined by a few critical moments around contact. In FACTR 2, we use force to find these moments and prioritize them for training. We find this helps policies learn better alignment and recovery behaviors, like the example below. w/ Steven Oh Jason Liu 🧵(1/N)

What if some parts of a robot demonstration are more important than others? Most of a trajectory is free-space motion. But success or failure is often determined by a few critical moments around contact. In FACTR 2, we use force to find these moments and prioritize them for training. We find this helps policies learn better alignment and recovery behaviors, like the example below. w/ Steven Oh Jason Liu 🧵(1/N)

Tony Tao

24,025 次观看 • 1 个月前

Force-sensing fingers! 🧤 Stanford researchers just released UMI-FT, a handheld data collection platform that puts compact six-axis force/torque sensors on each finger, enabling finger-level wrench measurements alongside RGB, depth, and pose data. Many manipulation tasks require careful force modulation: too little force and the task fails, too much and you cause damage. But commercial force/torque sensors are expensive, bulky, and fragile, which has limited large-scale force-aware policy learning. UMI-FT changes the economics. The platform uses an iPhone for RGB vision, ultrawide RGB, depth, and pose via ARKit, with each finger sensorized using a CoinFT sensor to capture per-finger wrench information during manipulation. This multimodal data trains an adaptive compliance policy that predicts position targets, grasp force, and stiffness for execution on standard compliance controllers. The learned policy runs slowest and generates reference targets, while model-based compliance and force controllers provide delicate 6D compliance control and real-time force modulation. They tested on three contact-rich, force-sensitive tasks: whiteboard wiping (locate eraser, grasp, wipe until clean), skewering zucchini (grasp slice firmly, push onto stick until punctured), and lightbulb insertion (grasp bulb, align bayonet pin with socket slit, insert while overcoming spring force, rotate to light up). The results are clear. Policies without compliance struggle to modulate contact force and trigger safety faults from excessive force. Policies without force sensing fail to grasp unseen objects or resist reaction forces, causing slippage. Here's the project page: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Force-sensing fingers! 🧤 Stanford researchers just released UMI-FT, a handheld data collection platform that puts compact six-axis force/torque sensors on each finger, enabling finger-level wrench measurements alongside RGB, depth, and pose data. Many manipulation tasks require careful force modulation: too little force and the task fails, too much and you cause damage. But commercial force/torque sensors are expensive, bulky, and fragile, which has limited large-scale force-aware policy learning. UMI-FT changes the economics. The platform uses an iPhone for RGB vision, ultrawide RGB, depth, and pose via ARKit, with each finger sensorized using a CoinFT sensor to capture per-finger wrench information during manipulation. This multimodal data trains an adaptive compliance policy that predicts position targets, grasp force, and stiffness for execution on standard compliance controllers. The learned policy runs slowest and generates reference targets, while model-based compliance and force controllers provide delicate 6D compliance control and real-time force modulation. They tested on three contact-rich, force-sensitive tasks: whiteboard wiping (locate eraser, grasp, wipe until clean), skewering zucchini (grasp slice firmly, push onto stick until punctured), and lightbulb insertion (grasp bulb, align bayonet pin with socket slit, insert while overcoming spring force, rotate to light up). The results are clear. Policies without compliance struggle to modulate contact force and trigger safety faults from excessive force. Policies without force sensing fail to grasp unseen objects or resist reaction forces, causing slippage. Here's the project page: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

12,822 次观看 • 6 个月前

Open-source dexterous hands with fingertip sensors! 🪬 ORCA Dexterity just released three dexterous hand models. Their mission: is to democratize robotic hand dexterity. They're sharing progress on exceptional, low-cost hardware and a software layer from low-level control to robotic hand learning. This is the open-source approach to dexterous manipulation. It has 83 taxels per finger with 0.1 N force detection which pretty is impressive for an open-source design. Tactile sensing is critical for dexterous manipulation, knowing contact forces enables gentle grasping, slip detection, and force-controlled assembly. Also 700g weight for the lite version makes it practical for mounting on robot arms without exceeding payload limits. Lower weight means faster movements and lower torque requirements. Open hardware accelerates robotics by letting researchers and builders modify designs for their specific needs without starting from scratch! ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Open-source dexterous hands with fingertip sensors! 🪬 ORCA Dexterity just released three dexterous hand models. Their mission: is to democratize robotic hand dexterity. They're sharing progress on exceptional, low-cost hardware and a software layer from low-level control to robotic hand learning. This is the open-source approach to dexterous manipulation. It has 83 taxels per finger with 0.1 N force detection which pretty is impressive for an open-source design. Tactile sensing is critical for dexterous manipulation, knowing contact forces enables gentle grasping, slip detection, and force-controlled assembly. Also 700g weight for the lite version makes it practical for mounting on robot arms without exceeding payload limits. Lower weight means faster movements and lower torque requirements. Open hardware accelerates robotics by letting researchers and builders modify designs for their specific needs without starting from scratch! ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

38,094 次观看 • 4 个月前

There are valuable lessons to be learned from South Africa. The Boers didn't lose control of their country because they lacked boots on the ground. They lost it because they failed to win the hearts and minds of the people. We need to avoid falling into the same trap by staying fully engaged on two fronts: 1. The propaganda front: exposing the enemy, re-energizing nationalistic spirit, mobilizing our people, building networks, uniting all Whites, and reaching critical mass. 2. The readiness front: forming organizations, training, confronting traitors and invaders in the streets, developing capabilities outside the system, and backing our political demands with force. Securing one without the other won't be enough to turn the ship around. We need to dominate on every level.

There are valuable lessons to be learned from South Africa. The Boers didn't lose control of their country because they lacked boots on the ground. They lost it because they failed to win the hearts and minds of the people. We need to avoid falling into the same trap by staying fully engaged on two fronts: 1. The propaganda front: exposing the enemy, re-energizing nationalistic spirit, mobilizing our people, building networks, uniting all Whites, and reaching critical mass. 2. The readiness front: forming organizations, training, confronting traitors and invaders in the streets, developing capabilities outside the system, and backing our political demands with force. Securing one without the other won't be enough to turn the ship around. We need to dominate on every level.

LIMES

95,182 次观看 • 6 个月前

General "Raizin" Caine, details the precision extraction mission in Venezuela: "Over the course of the night, aircraft began launching from 20 different bases on land and sea across the Western Hemisphere. In total, more than 150 aircraft, bombers, fighters, intelligence, reconnaissance, surveillance, rotary wing, were in the air last night. "Our youngest crew member was 20 and our oldest crew member was 49. And there's simply no match for American military might. As the night began, the helicopters took off with the extraction force, which included law enforcement officers, and began their flight into Venezuela at 100 feet above the water. As they approached Venezuelan shores, the United States began layering different effects provided by SPACECOM, CYBERCOM, and other members of the inter-agency to create a pathway. "Overhead, those forces were protected from aircraft- were protected by aircraft from the United States Marines, the United States Navy, the United States Air Force, and the Air National Guard. The force included F-22s, F-35s, F-18s, EA-18s, E-2s, B-1 bombers, and other support aircraft, as well as numerous remotely piloted drones. As the force began to approach Caracas, the joint air component began dismantling and disabling the air defense systems in Venezuela, employing weapons to ensure the safe passage of the helicopters into the target area. "The goal of our air component is, was, and always will be to protect the helicopters and the ground force, and get them to the target, and get them home. As the force crossed the last point of high terrain where they'd been hiding in the clutter, we assessed that we had maintained totally the element of surprise. As the helicopter force ingressed towards the objective at low level, we arrived at Maduro's compound at 1:01 AM Eastern Standard Time, or 2:01 AM Caracas local time, and the apprehension force descended into Maduro's compound and moved with speed, precision, and discipline towards their objective, and isolated the area to ensure the safety and security of the ground force while apprehending the indicted persons. "On arrival into the target area, the helicopters came under fire and they replied with that fire with overwhelming force in self-defense. One of our aircraft was hit, but remained flyable, and as the President said earlier today, all of our aircraft came home. And that aircraft remained flyable during the rest of the mission. As the operation unfolded at the compound, our air and ground intelligence teams provided real-time updates to the ground force, ensuring those forces could safely navigate the complex environment without unnecessary risk. "The force remained protected by overhead tactical aviation. Maduro and his wife, both indicted, gave up and were taken into custody by the Department of Justice, assisted by our incredible US military with professionalism and precision, with, with no loss of US life. After securing the indicted persons, the force began to prep for departure."

General "Raizin" Caine, details the precision extraction mission in Venezuela: "Over the course of the night, aircraft began launching from 20 different bases on land and sea across the Western Hemisphere. In total, more than 150 aircraft, bombers, fighters, intelligence, reconnaissance, surveillance, rotary wing, were in the air last night. "Our youngest crew member was 20 and our oldest crew member was 49. And there's simply no match for American military might. As the night began, the helicopters took off with the extraction force, which included law enforcement officers, and began their flight into Venezuela at 100 feet above the water. As they approached Venezuelan shores, the United States began layering different effects provided by SPACECOM, CYBERCOM, and other members of the inter-agency to create a pathway. "Overhead, those forces were protected from aircraft- were protected by aircraft from the United States Marines, the United States Navy, the United States Air Force, and the Air National Guard. The force included F-22s, F-35s, F-18s, EA-18s, E-2s, B-1 bombers, and other support aircraft, as well as numerous remotely piloted drones. As the force began to approach Caracas, the joint air component began dismantling and disabling the air defense systems in Venezuela, employing weapons to ensure the safe passage of the helicopters into the target area. "The goal of our air component is, was, and always will be to protect the helicopters and the ground force, and get them to the target, and get them home. As the force crossed the last point of high terrain where they'd been hiding in the clutter, we assessed that we had maintained totally the element of surprise. As the helicopter force ingressed towards the objective at low level, we arrived at Maduro's compound at 1:01 AM Eastern Standard Time, or 2:01 AM Caracas local time, and the apprehension force descended into Maduro's compound and moved with speed, precision, and discipline towards their objective, and isolated the area to ensure the safety and security of the ground force while apprehending the indicted persons. "On arrival into the target area, the helicopters came under fire and they replied with that fire with overwhelming force in self-defense. One of our aircraft was hit, but remained flyable, and as the President said earlier today, all of our aircraft came home. And that aircraft remained flyable during the rest of the mission. As the operation unfolded at the compound, our air and ground intelligence teams provided real-time updates to the ground force, ensuring those forces could safely navigate the complex environment without unnecessary risk. "The force remained protected by overhead tactical aviation. Maduro and his wife, both indicted, gave up and were taken into custody by the Department of Justice, assisted by our incredible US military with professionalism and precision, with, with no loss of US life. After securing the indicted persons, the force began to prep for departure."

Andrew Kolvet

76,491 次观看 • 7 个月前

Physics feels stable, predictable, and well-behaved largely because we learned it in 3D. That comfort hides a trap. In 1907, Paul Ehrenfest pointed out something unsettling. If you take the laws we treat as fundamental and transplant them into a different number of spatial dimensions, they often stop working the way we expect. Not just numerically different but qualitatively different. The issue isn’t the force law by itself. It’s geometry. Gauss’s law ties inverse-square forces to the surface area of spheres, and sphere geometry depends on dimension. Change the dimension, and the same-looking force produces a different potential, a different balance of attraction and inertia, and a different fate for motion. You can see this cleanly with a single problem of central force motion. In d spatial dimensions, flux conservation gives F(r) ∝ 1 / rᵈ⁻¹ so the potential scales as V(r) ∝ −1 / rᵈ⁻² (for d ≠ 2) Now add angular momentum. The effective radial potential becomes V_eff(r) = L² / (2 m r²) − C / rᵈ⁻² In 3D, those two terms balance in just the right way to allow stable bound orbits. Small perturbations stay small. Atoms don’t collapse. Planets don’t spiral away. In other dimensions, that balance breaks. In 2D, the force becomes 1/r, the potential becomes logarithmic, and bound motion sits on a knife edge. In 4D and higher, the attractive term becomes too steep. The centrifugal barrier loses the fight. Orbits plunge or escape. Same equations. Same initial conditions. Different dimension. Different physics. This isn’t science fiction. It’s a warning label. So it's clear that a lot of what we call physical intuition is really three-dimensional intuition wearing a lab coat. #Physics #MathematicalPhysics #ClassicalMechanics #DynamicalSystems #Geometry #Ehrenfest

Physics feels stable, predictable, and well-behaved largely because we learned it in 3D. That comfort hides a trap. In 1907, Paul Ehrenfest pointed out something unsettling. If you take the laws we treat as fundamental and transplant them into a different number of spatial dimensions, they often stop working the way we expect. Not just numerically different but qualitatively different. The issue isn’t the force law by itself. It’s geometry. Gauss’s law ties inverse-square forces to the surface area of spheres, and sphere geometry depends on dimension. Change the dimension, and the same-looking force produces a different potential, a different balance of attraction and inertia, and a different fate for motion. You can see this cleanly with a single problem of central force motion. In d spatial dimensions, flux conservation gives F(r) ∝ 1 / rᵈ⁻¹ so the potential scales as V(r) ∝ −1 / rᵈ⁻² (for d ≠ 2) Now add angular momentum. The effective radial potential becomes V_eff(r) = L² / (2 m r²) − C / rᵈ⁻² In 3D, those two terms balance in just the right way to allow stable bound orbits. Small perturbations stay small. Atoms don’t collapse. Planets don’t spiral away. In other dimensions, that balance breaks. In 2D, the force becomes 1/r, the potential becomes logarithmic, and bound motion sits on a knife edge. In 4D and higher, the attractive term becomes too steep. The centrifugal barrier loses the fight. Orbits plunge or escape. Same equations. Same initial conditions. Different dimension. Different physics. This isn’t science fiction. It’s a warning label. So it's clear that a lot of what we call physical intuition is really three-dimensional intuition wearing a lab coat. #Physics #MathematicalPhysics #ClassicalMechanics #DynamicalSystems #Geometry #Ehrenfest

Mathelirium

32,138 次观看 • 5 个月前

AgiBot’s new generation of industrial-grade interactive embodied robot, AgiBot G2, has officially launched! The G2 has already secured orders worth hundreds of millions of RMB, including two separate contracts each exceeding 100 million RMB, and has begun its first commercial deliveries. The AgiBot G2 is built to industrial standards, featuring high-performance joints, precision torque sensors, and an advanced spatial perception system. It supports rapid learning and deployment, offers strong multimodal voice interaction, and is designed for general use in industrial, logistics, and guidance scenarios. Inheriting the successful "Collect-Train-Deploy" model of its predecessor, the G1, the G2 brings significant upgrades, including a high-performance AI computing platform and actuators that enable omnidirectional obstacle avoidance and high-precision force-control tasks. Its 3-DOF waist allows for human-like bending and lateral body movement. A key feature is the G2's globally first-of-its-kind cross-shaped wrist force-control arm, which uses precision joint torque sensors and joint impedance control to delicately perceive external forces and respond smoothly. For continuous operation, the G2 supports autonomous charging and features a dual-battery hot-swapping system, meeting the 24-hour cycle demands of factory production lines. During the launch event, AgiBot demonstrated the G2’s ultra-low latency remote operation (teleoperation) capabilities. Operators successfully demonstrated precision shots (like hitting a floating balloon in Shanghai while operating from Beijing), showcasing the robot's high accuracy and low latency in both line-of-sight and beyond-line-of-sight scenarios. The G2 is already being deployed across four key real-world scenarios: In automotive parts production, it assists humans with tasks like safety belt lock core pressing and material handling. In precision operations, it used reinforcement learning to master delicate tasks like inserting memory sticks in just one hour. In logistics, the G2, enhanced by AgiBot's OmniHand dexterous hand, efficiently handles various package types for sorting and loading. Its strong mobility allows it to adapt to over 95% of factory floors. AgiBot is also commencing the first batch of commercial deliveries under an over 100 million RMB procurement contract with Joyson Electronic, formally landing the G2 in the automotive parts manufacturing sector.

AgiBot’s new generation of industrial-grade interactive embodied robot, AgiBot G2, has officially launched! The G2 has already secured orders worth hundreds of millions of RMB, including two separate contracts each exceeding 100 million RMB, and has begun its first commercial deliveries. The AgiBot G2 is built to industrial standards, featuring high-performance joints, precision torque sensors, and an advanced spatial perception system. It supports rapid learning and deployment, offers strong multimodal voice interaction, and is designed for general use in industrial, logistics, and guidance scenarios. Inheriting the successful "Collect-Train-Deploy" model of its predecessor, the G1, the G2 brings significant upgrades, including a high-performance AI computing platform and actuators that enable omnidirectional obstacle avoidance and high-precision force-control tasks. Its 3-DOF waist allows for human-like bending and lateral body movement. A key feature is the G2's globally first-of-its-kind cross-shaped wrist force-control arm, which uses precision joint torque sensors and joint impedance control to delicately perceive external forces and respond smoothly. For continuous operation, the G2 supports autonomous charging and features a dual-battery hot-swapping system, meeting the 24-hour cycle demands of factory production lines. During the launch event, AgiBot demonstrated the G2’s ultra-low latency remote operation (teleoperation) capabilities. Operators successfully demonstrated precision shots (like hitting a floating balloon in Shanghai while operating from Beijing), showcasing the robot's high accuracy and low latency in both line-of-sight and beyond-line-of-sight scenarios. The G2 is already being deployed across four key real-world scenarios: In automotive parts production, it assists humans with tasks like safety belt lock core pressing and material handling. In precision operations, it used reinforcement learning to master delicate tasks like inserting memory sticks in just one hour. In logistics, the G2, enhanced by AgiBot's OmniHand dexterous hand, efficiently handles various package types for sorting and loading. Its strong mobility allows it to adapt to over 95% of factory floors. AgiBot is also commencing the first batch of commercial deliveries under an over 100 million RMB procurement contract with Joyson Electronic, formally landing the G2 in the automotive parts manufacturing sector.

RoboHub🤖

33,831 次观看 • 9 个月前

China’s pretty humanoid robot stuns by opening a car door in a ‘world’s first’ | Jijo Malayil, Interesting Engineering Mornine used onboard sensors and full-body control to locate the handle, adjust posture, and open a car door—no human input needed. AiMOGA Robotics has claimed to have reached a significant milestone in embodied AI with its humanoid robot, Mornine, autonomously opening a car door inside a functioning Chery dealership in China. Relying solely on onboard sensors, full-body motion control, and end-to-end reinforcement learning, Mornine performed the task without any human input. Unlike scripted or teleoperated robots, Mornie identified the door handle, adjusted its posture, and used coordinated force across its limbs and torso to complete the action—demonstrating advanced autonomy in a real-world setting. “The deployment marks one of the first instances of a service robot executing such a high-friction, physical interaction in a live commercial setting,” said the firm in a statement. In April, at the Shanghai Auto Show, automotive brands Omoda and Jaecoo, subsidiaries of Chery Automobile, introduced Mornine, designed for use in car dealerships. From sim to service Opening a car door may seem like a simple task, but AiMOGA Robotics views it as a pivotal moment in robotics—signaling a shift from simulation to real-world service, and from basic command execution to autonomous capability. Using only onboard sensors and full-body motion control, Mornine identified the door handle, adjusted her posture, and applied coordinated force across her limbs to open the door—entirely without human intervention. Mornine’s advanced sensor suite includes 3D LiDAR, depth and wide-angle cameras, and a visual-language model (VLM), enabling real-time perception of door position and opening status. Uniquely, Mornine wasn’t explicitly programmed to recognize door handles. Instead, she learned through reinforcement learning, undergoing millions of simulated cycles to focus on the right region and perform the task independently. “We never explicitly told the robot what a door handle is. It learned to focus on that region by itself,” said the engineering team at AiMOGA Robotics in a statement. The learned model was transferred to the real world using Sim2Real methods. Mornine continuously gathers live sensor data during operation, which feeds into a cloud-based training loop, allowing her to improve through continuous learning in real-world settings, reports Robotics Tomorrow. Now active in multiple Chery 4S dealerships in China, Mornine not only opens car doors but also assists with customer greetings, vehicle introductions, and item delivery—marking a step forward in humanoid robotics for commercial retail environments. AI meets retail Originally introduced as the AiMOGA Robot, Mornine was developed to support dealership sales by performing tasks such as explaining vehicle specifications, leading showroom tours, serving refreshments, and engaging with customers in multiple languages. First conceived by Chery as a virtual character to appeal to Generation Z using metaverse and virtual human technologies, Mornine gradually evolved into a real-world interactive humanoid. After multiple iterations of character and model design, Mornine debuted as a digital persona in animations, livestreams, and promotional content, gaining brand recognition. Chery later expanded the concept beyond the virtual space, resulting in the creation of the AiMOGA humanoid robot. Leveraging Chery’s expertise in autonomous driving, environmental sensing, and control systems, AiMOGA features full-stack capabilities in perception, cognition, decision-making, and execution. It uses multimodal sensing—combining speech, vision, and environmental data—to interpret user gestures, commands, and showroom dynamics. A bionic motion system and automotive-grade hardware enable dexterous movement and upright mobility, while multi-robot collaboration allows for coordinated tasks like guided tours. At the decision-making layer, Deepseek’s large language models enable natural language understanding and personalized interaction. In April 2025, Mornine officially began commercial service as an “Intelligent Sales Consultant” at the OMODA C5 JOYSTAR 4S dealership in Kuala Lumpur, Malaysia—marking her full transition from a virtual concept to a real-world humanoid sales assistant.

Owen Gregorian

67,975 次观看 • 1 年前

U.S. based humanoid robotics company 1X has just unveiled their new tendon-driven robot hands with 25 degrees of freedom (DOF). • Made in USA • Tendon Drive Ratio: 5:1–15:1 • Wrist Dexterity: 3 DOF • Backdrivability: Fully backdrivable • Tactile Sensing: Pressure + location + slip • Finger Force: Up to 45 N • Wrist Torque: 17.75 Nm • Position Accuracy: ±0.2 mm • Waterproof Rating: IP68 • Reliability: >2 million cycles "These hands are designed to do something fundamental: remove the hardware ceiling on what humanoid robots can actually do, and make data the only barrier to capabilities. By matching or surpassing human hands across the dimensions that matter, they ensure our AI models are no longer limited by dexterity. NEO can now perform virtually any task a human can do with their hands– with the precision, adaptability, and gentleness required for real-world environments."

U.S. based humanoid robotics company 1X has just unveiled their new tendon-driven robot hands with 25 degrees of freedom (DOF). • Made in USA • Tendon Drive Ratio: 5:1–15:1 • Wrist Dexterity: 3 DOF • Backdrivability: Fully backdrivable • Tactile Sensing: Pressure + location + slip • Finger Force: Up to 45 N • Wrist Torque: 17.75 Nm • Position Accuracy: ±0.2 mm • Waterproof Rating: IP68 • Reliability: >2 million cycles "These hands are designed to do something fundamental: remove the hardware ceiling on what humanoid robots can actually do, and make data the only barrier to capabilities. By matching or surpassing human hands across the dimensions that matter, they ensure our AI models are no longer limited by dexterity. NEO can now perform virtually any task a human can do with their hands– with the precision, adaptability, and gentleness required for real-world environments."

Sawyer Merritt

742,478 次观看 • 23 天前

Why did Pakistan attack Afghanistan?! Pakistan initiated the attack on Afghanistan in an attempt: 1- To export its internal problems and rally the oppressed public behind its unpopular regime (there are dozens of active anti-regime militias). 2- To carry out external agendas, the assault on Afghanistan came right after Trump’s threats. 3- And due to the opposing ideologies between the leadership of Afghanistan (conservative Islam) and the secular Pakistani regime. All of these proxy regimes see a conservative Muslim government as a threat, aligning with their Western backers. The UAE and Saudi Arabia had no border disputes with Egypt, Tunisia, Libya, or Sudan, yet they did to those countries exactly what the Pakistani regime is now trying to do to Afghanistan. The Pakistani military regime’s actions will lead to more division within Pakistan. The only solution for Pakistan is to wake up and resolve these issues by sitting with all the opposing armed militias and listening to their demands, whether cultural, linguistic, or religious, and by allowing freedom of speech for everyone. Force will only create more division. The Western establishment’s goal is the division of Pakistan. The biggest threat to Pakistan is not India or any external enemy, it’s internal. And the biggest threat to them is a non-proxy Muslim regime with nukes. A nuclear power cannot be destroyed by an external force; it collapses from within. Look at the USSR. First Afghanistan, then bankruptcy, then division. #VIDEO1and2 the very militias the Pakistani regime claims to be fighting in Afghanistan roaming freely in Pakistani streets. #VIDEO3, released after the first strikes on Kabul, which the Pakistani regime’s media claimed targeted TTP leader Mehsud, refuting the claim, confirming that he is neither in Afghanistan nor killed.

Why did Pakistan attack Afghanistan?! Pakistan initiated the attack on Afghanistan in an attempt: 1- To export its internal problems and rally the oppressed public behind its unpopular regime (there are dozens of active anti-regime militias). 2- To carry out external agendas, the assault on Afghanistan came right after Trump’s threats. 3- And due to the opposing ideologies between the leadership of Afghanistan (conservative Islam) and the secular Pakistani regime. All of these proxy regimes see a conservative Muslim government as a threat, aligning with their Western backers. The UAE and Saudi Arabia had no border disputes with Egypt, Tunisia, Libya, or Sudan, yet they did to those countries exactly what the Pakistani regime is now trying to do to Afghanistan. The Pakistani military regime’s actions will lead to more division within Pakistan. The only solution for Pakistan is to wake up and resolve these issues by sitting with all the opposing armed militias and listening to their demands, whether cultural, linguistic, or religious, and by allowing freedom of speech for everyone. Force will only create more division. The Western establishment’s goal is the division of Pakistan. The biggest threat to Pakistan is not India or any external enemy, it’s internal. And the biggest threat to them is a non-proxy Muslim regime with nukes. A nuclear power cannot be destroyed by an external force; it collapses from within. Look at the USSR. First Afghanistan, then bankruptcy, then division. #VIDEO1and2 the very militias the Pakistani regime claims to be fighting in Afghanistan roaming freely in Pakistani streets. #VIDEO3, released after the first strikes on Kabul, which the Pakistani regime’s media claimed targeted TTP leader Mehsud, refuting the claim, confirming that he is neither in Afghanistan nor killed.

Warfare Analysis

45,598 次观看 • 9 个月前

AgiBot has formally unveiled its G2 humanoid robot, a system designed to transition into various industries and liberate humans from repetitive labor. G2 features high-performance joints, precision torque sensors, and an advanced spatial perception system, supporting quick deployment and multi-modal voice interaction. ► Factory Floor Performance: The G2 is engineered to industrial standards. In a safety belt lock production line, robots collaborate with human workers, performing tasks like pressing lock cores. The G2 collects production data to continuously train and iterate models (local server deployment ensures data privacy), steadily improving its operational ability. ► Mobility & Safety: The G2 navigates narrow factory aisles using dual LiDAR and full-panorama vision for environment sensing and collision detection. Its chassis is designed to overcome common obstacles (speed bumps, elevator gaps). It supports 24/7 continuous operation via autonomous return-to-charge and battery swapping. ► Humanoid Design Advantage: The G2's design includes a three-degree-of-freedom flexible waist, allowing it to mimic natural human movements like bending and side-leaning. This dramatically expands its operational workspace and enables seamless integration into existing human-centric production lines without costly modifications. ► Advanced Dexterity & Learning (Lab): The new G02 arm is the world's first cross-moment arm, featuring high-precision joint torque sensors that allow it to precisely sense external forces and adjust stiffness, mimicking human hand compliance. Using Real-Machine Reinforcement Learning (RL), the G2 can learn complex, delicate tasks like memory stick insertion in about one hour with minimal human intervention. ► Logistics & Grasping: In logistics sorting, the G2 uses a 19-degree-of-freedom mechanical dexterous hand (20N maximum fingertip force; 35kg capacity for hard objects) equipped with 3D tactile sensors to ensure it grasps securely without damaging items. Its full-body articulation (waist and legs) aids grasping and posture adjustment. ► Model & Data: G2's intelligence is powered by the Go-One Large Embodied Model (VLA architecture: Vision-Language-Latent Action) and the GE-One World Model (vision-centric predictive modeling), trained using the AgiBot Word true-machine dataset (over 500k downloads). ► Service & Interaction: The G2 is deployed as a guide/receptionist in settings like art museums. It uses its high-DOF head, arms, and waist to point to exhibits, maintains eye contact while navigating difficult spaces (chassis walks forward, body faces backward), handles specialized and random queries, and uses proactive safety features (stops movement, issues warnings) when people get too close.

AgiBot has formally unveiled its G2 humanoid robot, a system designed to transition into various industries and liberate humans from repetitive labor. G2 features high-performance joints, precision torque sensors, and an advanced spatial perception system, supporting quick deployment and multi-modal voice interaction. ► Factory Floor Performance: The G2 is engineered to industrial standards. In a safety belt lock production line, robots collaborate with human workers, performing tasks like pressing lock cores. The G2 collects production data to continuously train and iterate models (local server deployment ensures data privacy), steadily improving its operational ability. ► Mobility & Safety: The G2 navigates narrow factory aisles using dual LiDAR and full-panorama vision for environment sensing and collision detection. Its chassis is designed to overcome common obstacles (speed bumps, elevator gaps). It supports 24/7 continuous operation via autonomous return-to-charge and battery swapping. ► Humanoid Design Advantage: The G2's design includes a three-degree-of-freedom flexible waist, allowing it to mimic natural human movements like bending and side-leaning. This dramatically expands its operational workspace and enables seamless integration into existing human-centric production lines without costly modifications. ► Advanced Dexterity & Learning (Lab): The new G02 arm is the world's first cross-moment arm, featuring high-precision joint torque sensors that allow it to precisely sense external forces and adjust stiffness, mimicking human hand compliance. Using Real-Machine Reinforcement Learning (RL), the G2 can learn complex, delicate tasks like memory stick insertion in about one hour with minimal human intervention. ► Logistics & Grasping: In logistics sorting, the G2 uses a 19-degree-of-freedom mechanical dexterous hand (20N maximum fingertip force; 35kg capacity for hard objects) equipped with 3D tactile sensors to ensure it grasps securely without damaging items. Its full-body articulation (waist and legs) aids grasping and posture adjustment. ► Model & Data: G2's intelligence is powered by the Go-One Large Embodied Model (VLA architecture: Vision-Language-Latent Action) and the GE-One World Model (vision-centric predictive modeling), trained using the AgiBot Word true-machine dataset (over 500k downloads). ► Service & Interaction: The G2 is deployed as a guide/receptionist in settings like art museums. It uses its high-DOF head, arms, and waist to point to exhibits, maintains eye contact while navigating difficult spaces (chassis walks forward, body faces backward), handles specialized and random queries, and uses proactive safety features (stops movement, issues warnings) when people get too close.

RoboHub🤖

46,733 次观看 • 9 个月前

OPERATION HADIN KAI FOILS MASS ABDUCTION ATTEMPT AT FGGC MONGUNO Troops of Operation HADIN KAI (OPHK), in collaboration with personnel of the Nigeria Police Mobile Force (MOPOL), successfully foiled an attempted mass abduction by ISWAP terrorists at the Federal Government Girls College (FGGC), Monguno, at about 0130 hours (1:30 a.m.) on 19 July 2026. The FGGC facility is currently being utilized by the Borno State Government as temporary hostel accommodation for students of the Federal Polytechnic, Monguno. The terrorists reportedly gained access to the facility with the assistance of suspected collaborators in an attempt to abduct students. Alert security personnel immediately engaged the terrorists with coordinated and overwhelming firepower, effectively stalling their advance with Sector 3 Quick Reaction Force (QRF) immediately reinforcing the school. Confronted by the superior combat capability and determined resistance of the security forces, the terrorists were forced to abandon their criminal mission and flee in confusion without achieving their objective. During the encounter, parts of the school infrastructure sustained damage but the attempt was well contained by troops in conjunction with the Nigeria Police Force personnel. Following the operation, troops successfully rescued and evacuated all 46 students to Kinnasara Barracks, Monguno, where they received immediate medical assessment and appropriate care. All rescued students have been confirmed medically stable and no student was abducted. Regrettably, some students were fatally struck by the sporadic gunfire from the terrorists during the firefight. Operation HADIN KAI extends its deepest condolences to the families of the deceased and reassures the public that all necessary measures are being taken to safeguard the lives of residents and protect critical public institutions across the North East in liaison with the Borno State Government. Exploitation of the incident is ongoing to identify and apprehend the suspected collaborators, while troops and other security agencies are actively tracking the fleeing terrorists. Operation HADIN KAI remains steadfast in its commitment to sustaining relentless pressure on terrorist elements, denying them freedom of action, dismantling their operational capability, and ensuring that educational institutions and other critical infrastructure across the North East remain safe and secure. MOHAMMED GONI Captain Acting Military Information Officer Headquarters Joint Task Force (North East) Operation HADIN KAI MAIDUGURI 19 July, 2026

OPERATION HADIN KAI FOILS MASS ABDUCTION ATTEMPT AT FGGC MONGUNO Troops of Operation HADIN KAI (OPHK), in collaboration with personnel of the Nigeria Police Mobile Force (MOPOL), successfully foiled an attempted mass abduction by ISWAP terrorists at the Federal Government Girls College (FGGC), Monguno, at about 0130 hours (1:30 a.m.) on 19 July 2026. The FGGC facility is currently being utilized by the Borno State Government as temporary hostel accommodation for students of the Federal Polytechnic, Monguno. The terrorists reportedly gained access to the facility with the assistance of suspected collaborators in an attempt to abduct students. Alert security personnel immediately engaged the terrorists with coordinated and overwhelming firepower, effectively stalling their advance with Sector 3 Quick Reaction Force (QRF) immediately reinforcing the school. Confronted by the superior combat capability and determined resistance of the security forces, the terrorists were forced to abandon their criminal mission and flee in confusion without achieving their objective. During the encounter, parts of the school infrastructure sustained damage but the attempt was well contained by troops in conjunction with the Nigeria Police Force personnel. Following the operation, troops successfully rescued and evacuated all 46 students to Kinnasara Barracks, Monguno, where they received immediate medical assessment and appropriate care. All rescued students have been confirmed medically stable and no student was abducted. Regrettably, some students were fatally struck by the sporadic gunfire from the terrorists during the firefight. Operation HADIN KAI extends its deepest condolences to the families of the deceased and reassures the public that all necessary measures are being taken to safeguard the lives of residents and protect critical public institutions across the North East in liaison with the Borno State Government. Exploitation of the incident is ongoing to identify and apprehend the suspected collaborators, while troops and other security agencies are actively tracking the fleeing terrorists. Operation HADIN KAI remains steadfast in its commitment to sustaining relentless pressure on terrorist elements, denying them freedom of action, dismantling their operational capability, and ensuring that educational institutions and other critical infrastructure across the North East remain safe and secure. MOHAMMED GONI Captain Acting Military Information Officer Headquarters Joint Task Force (North East) Operation HADIN KAI MAIDUGURI 19 July, 2026

Nigerian Army

19,180 次观看 • 13 天前

Milestone! We (robotic arms for gadgets assembly) finished the first commercial order, which brought the first revenue. Here are some learnings from this: The customer was a smart toy manufacturer. The task was to add a heatsink to Raspberry Pi. We received parts from them and returned the assembled modules back. Currently, it's done by teleoperation. Later it will be done by a remote employee via the Internet. Then it will be automated action by action, reducing the operator's time on this and making the task profitable. ps. If you have an assembly task that we can do for you asynchronically - leave a comment below. Learning 1. It's possible! This task which is usually done by the human arm with 5 fingers can be done with a two-finger gripper with the addition of a couple of simple tooling. The task was not simplified. We peeled off thin films from stickers, unpacked paper boxes, moved PCB boards full of components, etc. And no unsolvable problems have been encountered yet. Challenges: 1) The paper box shifted during the opening Solved with the plastic walls that you can lean against 2) Heat pad, stuck to the gripper instead of heat sync. Can be solved by gripper with a pump, but this time solved with the patience of the operator 3) The film on the pad is very thin. Turned out that sub-millimeter arm precision is enough to peel it off with just a regular gripper. 4) The working area has not enough space. You'll only know this by doing real tasks in bulk. This could be solved by an extra pair of long arms, but in this case, solved with the patience of the operator. I think that in the end, we will have 5-10 types of universal tooling and 5-10 types of grippers to solve almost all the problems in such assembly tasks. Learning 2. It's slow. It took 5 times more time, than doing it with human hands. But the good news is there's a lot of room for improvement. We now have specific “time for task” metrics, which we will decrease with iterations. The main reasons for slowness: 1) To rotate the gripper to a steep angle you are forced to control one robot arm with two hands instead of using both arms. We can fix this by just making more room for rotations. 2) Grabbing PCB board with two arms is hard. A slight difference in rotation can break the board, and it's hard to control these angles visually. To solve this, the best way is to use force feedback so you can feel the pressure applied to the item. 3) Accuracy and steadiness is still can be improved We will try a metal version and double the motors to do this. 4) It is physically difficult for the human hands to move with such precision To solve this, we will add a pad for the hands like in surgical robots Learning 3. It's a good business model The "Factory in the cloud" is a good business model for this stage. You send us parts and we send back assembled modules. Currently, it's more convenient than sending a robot to your place, as we can iterate/fix the robot quickly and utilize it 100% of the time. When we polish the set-up over time - we can send robots to your place. So if we can assemble something for you in the USA with Chinese prices by using modern automation - leave a comment below.

Milestone! We (robotic arms for gadgets assembly) finished the first commercial order, which brought the first revenue. Here are some learnings from this: The customer was a smart toy manufacturer. The task was to add a heatsink to Raspberry Pi. We received parts from them and returned the assembled modules back. Currently, it's done by teleoperation. Later it will be done by a remote employee via the Internet. Then it will be automated action by action, reducing the operator's time on this and making the task profitable. ps. If you have an assembly task that we can do for you asynchronically - leave a comment below. Learning 1. It's possible! This task which is usually done by the human arm with 5 fingers can be done with a two-finger gripper with the addition of a couple of simple tooling. The task was not simplified. We peeled off thin films from stickers, unpacked paper boxes, moved PCB boards full of components, etc. And no unsolvable problems have been encountered yet. Challenges: 1) The paper box shifted during the opening Solved with the plastic walls that you can lean against 2) Heat pad, stuck to the gripper instead of heat sync. Can be solved by gripper with a pump, but this time solved with the patience of the operator 3) The film on the pad is very thin. Turned out that sub-millimeter arm precision is enough to peel it off with just a regular gripper. 4) The working area has not enough space. You'll only know this by doing real tasks in bulk. This could be solved by an extra pair of long arms, but in this case, solved with the patience of the operator. I think that in the end, we will have 5-10 types of universal tooling and 5-10 types of grippers to solve almost all the problems in such assembly tasks. Learning 2. It's slow. It took 5 times more time, than doing it with human hands. But the good news is there's a lot of room for improvement. We now have specific “time for task” metrics, which we will decrease with iterations. The main reasons for slowness: 1) To rotate the gripper to a steep angle you are forced to control one robot arm with two hands instead of using both arms. We can fix this by just making more room for rotations. 2) Grabbing PCB board with two arms is hard. A slight difference in rotation can break the board, and it's hard to control these angles visually. To solve this, the best way is to use force feedback so you can feel the pressure applied to the item. 3) Accuracy and steadiness is still can be improved We will try a metal version and double the motors to do this. 4) It is physically difficult for the human hands to move with such precision To solve this, we will add a pad for the hands like in surgical robots Learning 3. It's a good business model The "Factory in the cloud" is a good business model for this stage. You send us parts and we send back assembled modules. Currently, it's more convenient than sending a robot to your place, as we can iterate/fix the robot quickly and utilize it 100% of the time. When we polish the set-up over time - we can send robots to your place. So if we can assemble something for you in the USA with Chinese prices by using modern automation - leave a comment below.

Igor Kulakov

37,266 次观看 • 1 年前

🚨 BREAKING: Microsoft's first robotics foundation model! 🤯 Microsoft just announced Rho-alpha (ρα), their first robotics model derived from the Phi series of vision-language models. Rho-alpha translates natural language commands into control signals for robotic systems performing bimanual manipulation tasks. Commands like "push the green button with the right gripper," "pull out the red wire," "flip the top switch on," or "turn the knob to position 5" get executed directly by dual-arm robots. What makes this different from standard vision-language-action (VLA) models is the additional modalities. Rho-alpha is a VLA+ model that adds tactile sensing to the perceptual mix, with plans to incorporate force feedback. On the learning side, the model is designed to continually improve during deployment by learning from human feedback. The training approach combines trajectories from physical demonstrations and simulated tasks with web-scale visual question answering data. Since teleoperation data is scarce and expensive, Microsoft is using NVIDIA Isaac Sim on Azure to generate physically accurate synthetic datasets via reinforcement learning. These simulated trajectories get combined with commercial and open physical demonstration datasets. The model is currently under evaluation on dual-arm setups and humanoid robots. Microsoft is opening an Early Access Program for organizations interested in evaluating Rho-alpha. Robots that can adapt to dynamic situations and human preferences are more useful in real environments and more trusted by the people operating them. Read more here: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

🚨 BREAKING: Microsoft's first robotics foundation model! 🤯 Microsoft just announced Rho-alpha (ρα), their first robotics model derived from the Phi series of vision-language models. Rho-alpha translates natural language commands into control signals for robotic systems performing bimanual manipulation tasks. Commands like "push the green button with the right gripper," "pull out the red wire," "flip the top switch on," or "turn the knob to position 5" get executed directly by dual-arm robots. What makes this different from standard vision-language-action (VLA) models is the additional modalities. Rho-alpha is a VLA+ model that adds tactile sensing to the perceptual mix, with plans to incorporate force feedback. On the learning side, the model is designed to continually improve during deployment by learning from human feedback. The training approach combines trajectories from physical demonstrations and simulated tasks with web-scale visual question answering data. Since teleoperation data is scarce and expensive, Microsoft is using NVIDIA Isaac Sim on Azure to generate physically accurate synthetic datasets via reinforcement learning. These simulated trajectories get combined with commercial and open physical demonstration datasets. The model is currently under evaluation on dual-arm setups and humanoid robots. Microsoft is opening an Early Access Program for organizations interested in evaluating Rho-alpha. Robots that can adapt to dynamic situations and human preferences are more useful in real environments and more trusted by the people operating them. Read more here: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

60,912 次观看 • 6 个月前

My ep w Allan Dafoe (director of frontier safety & governance at DeepMind): "Tech doesn't force us to do anything, it merely opens the door – and it's military-economic competition that forces us through." (32:25) "We're not at peak returns to generality." (1:38:06) "The counterposition I would offer is: you don't want to equip groups trying to shape history with a naive model of what's possible..." (20:31) "The agricultural revolution, evidence suggests, was not great for a lot of people... probably health and welfare went down." (13:27) "He gave a demonstration of cannons bombarding the shore... The Japanese asked him, 'Will you bring your ships again?' And he said, 'I'll bring more.'... That was the opening of Japan. It led to a 15-year period of revolution." (40:17) "I'm glad you think it's obvious. Let's not underestimate the bias that comes from a scientist using the tools that they prefer to use..." (33:30) "In my view, Demis and Shane are extremely impressive, from their safety orientation to… their wisdom and broad character. AGI is probably the most important historical development… Demis and Google DeepMind are very likely to be influential" (7:40) "...imagine we had no investment in AI alignment. Maybe it [only] delays it 2 years until the market demands we solve this problem..." (51:50) "We can talk about what I've called the 'super-cooperative AGI hypothesis': that as AI scales to AGI, so will cooperative competence scale to sort of infinity" (1:08:50) "Increasing the cooperative skill of AI will make those AI systems better off, but it may harm any agent who's excluded from that cooperative dynamic" (1:25:09) "...the Frontier Safety Framework I don't think is cheap talk. It really does represent an expression of Google DeepMind's sense of the nature of the risks..." (2:29:05) "...this is a new phenomenon, something new under the sun. It's quite incredible that finally the safety has reached the number of nines required..." (2:40:48) Links below, enjoy! 4:31 Why join Google DeepMind over everyone else? 9:30 Do humans control technological change? 38:11 Competition took away Japan's choice 1:03:34 How AI could boost cooperation between people and states 1:08:44 The super-cooperative AGI hypothesis 1:43:39 It matters what AGI learns first vs last 1:48:24 How Google tests for dangerous capabilities 2:00:57 Evals 'in the wild' 2:15:05 DeepMind's strategy for ensuring its frontier models don't cause harm 2:18:52 How 'structural risks' can force everyone into a worse world 2:29:01 How much do AI companies really want external regulation? 2:40:48 How AI could make life way better

My ep w Allan Dafoe (director of frontier safety & governance at DeepMind): "Tech doesn't force us to do anything, it merely opens the door – and it's military-economic competition that forces us through." (32:25) "We're not at peak returns to generality." (1:38:06) "The counterposition I would offer is: you don't want to equip groups trying to shape history with a naive model of what's possible..." (20:31) "The agricultural revolution, evidence suggests, was not great for a lot of people... probably health and welfare went down." (13:27) "He gave a demonstration of cannons bombarding the shore... The Japanese asked him, 'Will you bring your ships again?' And he said, 'I'll bring more.'... That was the opening of Japan. It led to a 15-year period of revolution." (40:17) "I'm glad you think it's obvious. Let's not underestimate the bias that comes from a scientist using the tools that they prefer to use..." (33:30) "In my view, Demis and Shane are extremely impressive, from their safety orientation to… their wisdom and broad character. AGI is probably the most important historical development… Demis and Google DeepMind are very likely to be influential" (7:40) "...imagine we had no investment in AI alignment. Maybe it [only] delays it 2 years until the market demands we solve this problem..." (51:50) "We can talk about what I've called the 'super-cooperative AGI hypothesis': that as AI scales to AGI, so will cooperative competence scale to sort of infinity" (1:08:50) "Increasing the cooperative skill of AI will make those AI systems better off, but it may harm any agent who's excluded from that cooperative dynamic" (1:25:09) "...the Frontier Safety Framework I don't think is cheap talk. It really does represent an expression of Google DeepMind's sense of the nature of the risks..." (2:29:05) "...this is a new phenomenon, something new under the sun. It's quite incredible that finally the safety has reached the number of nines required..." (2:40:48) Links below, enjoy! 4:31 Why join Google DeepMind over everyone else? 9:30 Do humans control technological change? 38:11 Competition took away Japan's choice 1:03:34 How AI could boost cooperation between people and states 1:08:44 The super-cooperative AGI hypothesis 1:43:39 It matters what AGI learns first vs last 1:48:24 How Google tests for dangerous capabilities 2:00:57 Evals 'in the wild' 2:15:05 DeepMind's strategy for ensuring its frontier models don't cause harm 2:18:52 How 'structural risks' can force everyone into a worse world 2:29:01 How much do AI companies really want external regulation? 2:40:48 How AI could make life way better

Rob Wiblin

89,899 次观看 • 1 年前

The Gerald R. Ford Carrier Strike Group is deploying next week to Europe. I took down an earlier post speculating about the composition. Here's what we know: • CNN reported today that GRFCSG "will likely move into the eastern Mediterranean Sea, near Israel, given the ongoing conflict." • Up to 6 destroyers will escort the world's largest, most lethal, and advanced aircraft carrier • Equipped with Anduril Roadrunner and Raytheon Coyote interceptor drone systems for the first time • Carrier Strike Group 12 (CSG-12) coordinates and directs the actions of USS Gerald R. Ford (CVN 78), USS Winston S. Churchill (DDG 81), Carrier Air Wing (CVW) 8 with its 9 squadrons, and Destroyer Squadron (DESRON) 2. (DVIDS, 11 April) • Ford and several escorts completed a 31-day COMPTUEX training in April, preparing for a potential conflict with Iran. COMPTUEX is the "most complex integrated training event and prepares naval task forces for sustained high-end Joint and combined combat." • On 6 May, destroyer USS Forrest Sherman (DDG 98) departed Norfolk as part of the GRFCSG for a scheduled deployment to U.S. 5th Fleet (DVIDS, 6 May) and currently in the Red Sea (USNI). U.S. Navy ships frequently disaggregate and operate independently under different command structures, such as 5th Fleet's Task Force 55. "As the first-in-class ship of Ford-class aircraft carriers, CVN 78 represents a generational leap in the U.S. Navy’s capacity to project power on a global scale." The flagship of CSG-12 "delivers the Joint force and Commander-in-Chief the combat capability to deter, and if necessary, defeat America's adversaries in support of national security and economic prosperity." More info, context, and sources below. Pre-deployment video released last month:

The Gerald R. Ford Carrier Strike Group is deploying next week to Europe. I took down an earlier post speculating about the composition. Here's what we know: • CNN reported today that GRFCSG "will likely move into the eastern Mediterranean Sea, near Israel, given the ongoing conflict." • Up to 6 destroyers will escort the world's largest, most lethal, and advanced aircraft carrier • Equipped with Anduril Roadrunner and Raytheon Coyote interceptor drone systems for the first time • Carrier Strike Group 12 (CSG-12) coordinates and directs the actions of USS Gerald R. Ford (CVN 78), USS Winston S. Churchill (DDG 81), Carrier Air Wing (CVW) 8 with its 9 squadrons, and Destroyer Squadron (DESRON) 2. (DVIDS, 11 April) • Ford and several escorts completed a 31-day COMPTUEX training in April, preparing for a potential conflict with Iran. COMPTUEX is the "most complex integrated training event and prepares naval task forces for sustained high-end Joint and combined combat." • On 6 May, destroyer USS Forrest Sherman (DDG 98) departed Norfolk as part of the GRFCSG for a scheduled deployment to U.S. 5th Fleet (DVIDS, 6 May) and currently in the Red Sea (USNI). U.S. Navy ships frequently disaggregate and operate independently under different command structures, such as 5th Fleet's Task Force 55. "As the first-in-class ship of Ford-class aircraft carriers, CVN 78 represents a generational leap in the U.S. Navy’s capacity to project power on a global scale." The flagship of CSG-12 "delivers the Joint force and Commander-in-Chief the combat capability to deter, and if necessary, defeat America's adversaries in support of national security and economic prosperity." More info, context, and sources below. Pre-deployment video released last month:

Ian Ellis

62,870 次观看 • 1 年前