Video Friday is your weekly selection of awesome robotics videos, collected by your friends at IEEE Spectrum robotics. We also post a weekly calendar of upcoming robotics events for the next few months. Please send us your events for inclusion. CoRL 2026: 9–12 November 2026, AUSTIN Humanoids 2026: 6–9 December 2026, SANTA CLARA, CA Enjoy today’s videos! This is such a novel and underrated form factor for a household robot. I absolutely love it. Stringman is all yours for US $795. [ Neufangled Robotics ] As of yesterday, Agility Robotics is publicly trading as AGLT. [ Agility Robotics ] Happy Halloween from Flexiv Robotics! Okay WHAT. [ Flexiv ] Nothing hides from SCUTL. Not even the paranormal. [ Ground Control Robotics ] Thanks, Dan! Happy Halloween from HEBI Robotics! We asked ourselves if you can carve a pumpkin using nothing but a T-Series arm and a drill originally designed to run an auger. The answer is yes, so long as you don’t care too much about accuracy. Basic pathing commands running off a laptop, stability provided by a quick connection to Treadward. Happy Halloween, HEBI fans![ HEBI ] Something strange has been happening at KIMLAB. Our snacks kept disappearing… and we finally caught the culprit. Meet GG, our new lab friend. GG is a character robot inspired by the Ggeomeoksali from Netflix’s The East Palace. GG sneaks around for snacks, eats them, sleeps most of the time, begs everyone for more food, and leans on whoever is nearby. Cute, comforting, and a good friend. But GG is still a gwi-mae… so who knows what else it can do. [ Kinetic Intelligent Machine Lab ] Happy Halloween from Dusty Robotics! [ Dusty ] Finally, a robot with a butt. [ JSK Lab ] Thanks, Shunnosuke! Any drone can crash into a tree once. This drone can crash into a tree MORE than once. [ ETH Zurich ] Lume is a robotic floor lamp that moves around your home, helps with everyday chores, and returns to being a beautiful lamp when it’s done. [ Syncere ] If we’re going to use robots at work or at home, they’re going to have to communicate with us. At Toyota Research Institute, a team of robotics user experience designers is working on the sounds, gestures, light patterns, and interfaces that allow robots to let us know that they are on the move, that we need to stay clear, or to tell us that they just really need help. Steffi Paepcke, Director of Robotics User Experience at TRI, accepts the challenge to explain how her team approaches this work in just one minute. I need a sound for when people around me need to give me space. [ Toyota Research Institute ] The entirely student-run, self-taught and highly collaborative MIT Combat Robotics Club (CRC) builds high energy fighting machines to compete in House Battles and larger competitions. Fans of the hit show BattleBots may remember some of CRC’s past robots, including Uppercut, SawBlaze, the Dentist, and Overhaul. CRC is open to every MIT student regardless of year, major, or prior experience. Students learn a variety of skills, from CAD and FEA to CNC Machining and Welding—and they learn through watching their creations get completely destroyed. [ MIT Combat Robotics Club ] Happy birthday JPL! Where do I get a shirt like that last guy has? [ JPL ] Can machines be creative? In this first episode of Moonshot, host Mike Massimino talks with Hod Lipson, a Columbia Engineering roboticist whose work includes robots that learn and adapt. Lipson, the James and Sally Scapa Professor of Innovation in Columbia’s Department of Mechanical Engineering, explains why physical AI is so difficult, how engineers and AI might work together on new designs, and how he defines creativity. He also shares advice on why playing it safe can limit your impact. [ Columbia University ] This closing fireside chat brings together two of the field’s most consequential builders—Jeff Dean, whose work at Google has helped define the modern AI stack, and Fei-Fei Li, whose case for human-centered AI has shaped how the field thinks about who this technology should actually serve. Coming at the end of a day spent grappling with AI’s risks and unresolved questions, the conversation turns to where the two of them agree, where they don’t, and what they think the choices ahead really come down to. [ Stanford Human-Centered AI ]