👀 Ever wondered what happens when you give a robot eyes and a brain? Well... I’ve been working on exactly that! 🤖✨ My new project (still under development) is an 𝗔𝗜 𝗩𝗶𝘀𝗶𝗼𝗻 𝗔𝘂𝘁𝗼𝗻𝗼𝗺𝗼𝘂𝘀 𝗥𝗼𝗯𝗼𝘁 that can see, detect, and react to its surroundings in real time. 🚀 The best part? Watching it recognize objects, make decisions on its own and navigating objects autonomously is so satisfying. 📹 Sharing a quick snippet of its AI vision in action — this is just the beginning! 🛠️ Tech stack powering it: 🔹𝗥𝗮𝘀𝗽𝗯𝗲𝗿𝗿𝘆 𝗣𝗶 𝟰𝗕 🔹𝗘𝗦𝗣𝟯𝟮 🔹𝗬𝗢𝗟𝗢𝘃𝟴-𝗟𝗶𝘁𝗲 𝗳𝗼𝗿 𝗔𝗜 𝗿𝗲𝗰𝗼𝗴𝗻𝗶𝘁𝗶𝗼𝗻 I want to thank my teammate Ishpreet Singh to add his valuable contributions to the project. Can’t wait to push this further — would love to hear your thoughts! 🙌 #AI #Robotics #ComputerVision #YOLOv8 #RaspberryPi #ESP32 #Innovation

AI researchers have tunnel vision and it's costing us the interface revolution. Every lab racing to build "world models" for robots is ignoring the world models we already inhabit — the ones flickering on your screen right now. 𝗧𝗵𝗶𝗻𝗸 𝗮𝗯𝗼𝘂𝘁 𝗶𝘁 𝘁𝗵𝗶𝘀 𝘄𝗮𝘆: When Gogol wrote "The Nose" — that amazing Russian story about a man whose nose walks around town conducting business — he created something profound. A world with its own 'physics' where the impossible becomes inevitable. Once you accept that noses can be bureaucrats, everything else makes perfect sense. 𝗧𝗵𝗲 𝗺𝗶𝘀𝘀𝗲𝗱 𝗼𝗽𝗽𝗼𝗿𝘁𝘂𝗻𝗶𝘁𝘆: We already generate interfaces by the terabyte through frameworks. The constraint systems exist. The physics are mappable. 2D interface generation should be the easier stepping stone to everything else. But instead? Radio silence from the research community. 𝗪𝗵𝗲𝗿𝗲'𝘀 𝗚𝗲𝗻𝗶𝗲 𝟰 𝗳𝗼𝗿 𝗶𝗻𝘁𝗲𝗿𝗳𝗮𝗰𝗲𝘀? Where's the system that watches how you work and generates the perfect UI for that exact moment? Why are we training models to understand object permanence for robots but not interface persistence for humans? This isn't just about better apps — it's about recognizing that world models are everywhere. Fiction, games, interfaces, robots — they're all constraint systems. They're all inhabitable worlds. They're all the same fundamental problem wearing different masks. 𝗛𝗲𝗿𝗲'𝘀 𝗺𝘆 𝗰𝗵𝗮𝗹𝗹𝗲𝗻𝗴𝗲 𝗳𝗼𝗿 𝘆𝗼𝘂: If you know any AI researchers, ask them this one question: "If you can build 3D world models for robots, why aren't you building them for 2D interfaces?" The first lab to connect these dots won't just improve how we use computers — they'll redefine what it means to interact with intelligence itself. We explore this deeper (including why your browser renders three Google apps but keeps them artificially separate): https://lnkd.in/gAMYJiT8 #AIResearch #WorldModels #InterfaceDesign #AI #MachineLearning #HumanComputerInteraction #TechInnovation

A* Search Algorithm in AI & Robotics 🧭🤖 When it comes to pathfinding and navigation, the A* (A-star) algorithm remains one of the most powerful and widely used techniques in Artificial Intelligence and Robotics. By combining the efficiency of Dijkstra’s algorithm with the foresight of heuristics, A* finds the shortest and most cost-effective path — making it ideal for robot navigation, game AI, and route optimization systems. Its balance between accuracy and performance makes it a cornerstone for intelligent movement and decision-making in dynamic environments. 👉 Have you implemented or worked with A* in your projects? 👉 What other pathfinding or planning algorithms do you find most effective for robotics applications? Let’s share experiences and insights on how algorithms like A* are shaping smarter, more adaptive machines. #AI #Robotics #PathPlanning #AStar #MachineLearning #Automation #Innovation #ArtificialIntelligence

🤖 Skild AI just unveiled a robot brain that refuses to quit. Shattered limbs? Jammed motors? Wrong body? Doesn’t matter. If the machine can move, the Skild Brain will move it, no matter what form it takes. This omni-bodied controller isn’t tied to a single robot. It can walk, roll, or crawl using whatever hardware it’s given. 🔥 How is this even possible? The Brain was trained for what amounts to 1,000 years in simulation, mastering 100,000 different robot bodies across countless virtual worlds. That training makes it one of the most adaptive AI locomotion systems ever built. Think back to that Terminator scene where half a robot keeps crawling forward after being blown apart. Skild’s breakthrough feels like a real-world step toward that level of resilience. The future of robotics isn’t about building the toughest body. It’s about building the smartest brain. 🔹 Follow AIPOOOL to Stay Ahead in the Race of Tech & AI 🚀 📩 Subscribe to our newsletter on LinkedIn 🔗 https://lnkd.in/gnUKzght #AI #Robotics #Innovation #Future #Simulation #Technology #Automation #MachineLearning #Resilience #Adaptability #Tech #News #AIPOOOL #Robots #Viral

AI Gets Street Smart Why can a 1979 Atari beat a state-of-the-art AI chatbot at chess? Because today’s AIs are “book smart,” not “street smart.” They are brilliant at predicting patterns from vast amounts of data, but they lack a true understanding of rules, physics or cause-and-effect. The chatbot loses because it’s guessing the next move, not reasoning about the game. The solution, and the next great leap in AI, is "world models". This approach trains AI in hyper-realistic simulations – like a pilot in a flight simulator – allowing it to learn from experience and build an internal model of reality. This is the key to unlocking "Physical AI" for robotics, self-driving cars and other autonomous systems that can operate in the real world. It marks the critical shift from AI that learns from static data to AI that learns from dynamic experience. Which do you think will ultimately be more powerful: an AI that has read everything ever written, or one that has experienced a million simulated realities? Source: The Wall Street Journal. #AI #ArtificialIntelligence #WorldModels #MachineLearning #FutureOfTech #Robotics

In the quest to automate artistic creation, have we inadvertently turned humans into robots? The pursuit of a science fiction future, where robots build everything, might be leading us to become overly reliant on algorithms. This raises a question about artificial intelligence and its impact on human autonomy. To what extent should we allow algorithms to dictate our actions and decisions? Find out more at https://togetherlearning.com/research/abyss #automation #artificialintelligence #humanautonomy #algorithms #futureofwork

Figure AI Inc. has been valued at $39 billion after securing massive funding. But what does this mean for the future of robotics and AI? Could this be the start of a tech revolution? 🔗 Full story in comments ↓ #RoboticsRevolution #AIInvestments #FigureAI #RoboticsFuture #TechValuations

Master AI and robotics with expert guidance! Turn curiosity into real-world innovation and equip young minds with the skills to lead tomorrow. Learn More: https://zurl.co/ECV0f #LearnAI #FutureInnovators #RoboticSkills #MilagrowEducation #innovation #stemlearning #edtech

Meet Unitree G1 a humanoid robot now pulling off Cristiano Ronaldo’s ‘Siuuu’ and LeBron James’ iconic moves. Thanks to advanced motion learning, robots are not just mimicking; they’re mastering agility, balance, and creativity once thought exclusive to humans. The future of sports and robotics is already here. #UnitreeG1 #HumanoidRobot #Siuuu #CristianoRonaldo #LeBronJames #IconicMoves #RobotDance #TechInnovation #AI

Huge announcement from Figure AI! Multitasking robots can now learn in human environments. This is a step towards robots working at homes helping in everyday tasks and supporting the elderlys independent life and well-being. Summary by copilot: Robots are learning from us—not just our commands, but our lived experience. Figure’s new initiative, #ProjectGoBig, marks a turning point in humanoid AI. By training the Helix model on egocentric human video, they’ve unlocked direct human-to-robot transfer: robots that can navigate and manipulate the world from natural language alone. “Go water the plants” is no longer a scripted task—it’s a learned behavior. This leap is powered by a bold partnership with Brookfield, whose vast portfolio of homes, offices, and logistics spaces will become real-world classrooms for embodied AI. It’s internet-scale pretraining, grounded in the complexity of human environments. As we enter this new phase of human-machine collaboration, the question isn’t just what can robots do?—but how do we guide what they learn? Leadership in the AI era means more than innovation. It means systems optimization with human dignity at the center. It means asking: 🔹 What values do we encode in our datasets? 🔹 How do we ensure wellbeing in machine-mediated spaces? 🔹 Who gets to shape the future of embodied intelligence? Let’s go big—but let’s go wisely. https://lnkd.in/drr7evuW #AILeadership #HumanCenteredTech #EmbodiedIntelligence #FigureAI #SystemsThinking #WellbeingByDesign