Figure AI's humanoid robots sort packages for 48 hours straight, then face off against human intern

The Internet can't stop watching Figure AI's humanoid robots handling packages

The robotics startup Figure AI has been livestreaming humanoid robots placing thousands of packages onto a conveyor belt for nearly a week -- a spectacle that included a robot competing against a human intern at one point. The promotional robot demo has become a viral sensation among tech enthusiasts, spurring YouTube commenters to name the robots and the company rapidly rolling out related robot merchandise in response. Users on X have described the livestream in glowing terms such as "the greatest product demo since Steve Jobs' 'one more thing.'" But despite such sentiments, it's worth bearing in mind that even the most impressive robot demos represent narrow windows for understanding real-world robot capabilities. Figure's event began on May 13 as a planned eight-hour robot demonstration featuring the company's latest Figure 03 robots. The chosen robotic task involved inspecting the barcodes on various small packages -- including cardboard boxes and soft padded envelopes or bags -- and then placing the packages on a conveyor belt with the barcodes facing downward. The demo would feature the robots performing the task autonomously without any human intervention, according to Figure CEO Brett Adcock. But Adock initially played down expectations by noting that the Figure team was aiming for the robots to work for eight hours straight, whereas a previous Figure demo had lasted just one hour. "High odds something breaks," Adcock posted on X. The robots rely on the company's Helix 02 neural network system that supposedly enables full-body control and "long horizon autonomy" to direct the robot's actions for various tasks. Figure's website describes the robots' whole-body controller system as having been trained on more than 1,000 hours of human motion data, along with spending time training in simulation across more than 200,000 parallel environments. That Helix 02 system runs "entirely onboard" each robot's hardware with AI inference being done on the device, Adcock explained in his X post. However, the robots are networked together for communication purposes, and so they can autonomously request another robot to step in if they need to recharge their batteries -- each robot is expected to work about three to four hours before its batteries run low. The robots may also swap out if they encounter hardware or software issues. It keeps going and going By the time Figure had livestreamed eight hours of the robots performing "autonomous, unsupervised work," Adcock was declaring that the team had decided to keep going with the livestream 24/7. He also highlighted YouTube comments that named several of the robots Bob, Frank, and Gary. On May 14, the robots had surpassed 30 hours of collective work, with individual robots taking turns swapping in and out. Adcock was capitalizing on the attention by wearing a T-shirt with the image of the robot dubbed "Frank," all while touting the company's merch store to viewers. He also welcomed another robot to the team by attaching a nametag with the name Rose. Adding to the spectacle, people began placing bets through the prediction market Polymarket on how long the robots could run without failure and how many packages they could handle. By May 15, the robots had seemingly achieved "48 hours of nonstop autonomous operation without a failure," Adcock posted on X. "We are now running this until a failure to perform the use case," he added. David McCall, Figure's head of design, also appeared briefly on the livestream to give another robot a "Jim" nametag. That particular robot would feature prominently in the Figure team's next attention-grabbing scheme in response to one viewer's comment -- pitting robot against human on the same task. Human versus machine On May 17, Adcock laid out the "Man vs. Machine" scenario for the head-to-head competition that would last 10 hours. The human competitor would get meal breaks and paid rest breaks during the shift in accordance with California labor laws. Aimé Gérard, an intern working at Figure AI, was chosen to represent the human side. A comparison video revealed some of the current differences between the humanoid robots' performance and human capabilities. Whereas Gérard could speedily and precisely pick up packages for inspection before moving them along, the robots moved at a slower but methodical pace. The robots also sometimes spent extra time trying and failing to pick up packages or grabbed repeatedly at empty air when trying to sweep packages along. That difference allowed Gérard to stay ahead on the package task until he took a mandated break, according to Business Insider. But he quickly recovered the human lead from the robotic competition once he returned -- eventually claiming victory by sorting 12,924 packages versus the robots' 12,732 packages. The human intern worked at a rate of 2.79 seconds per package, whereas the robots averaged 2.83 seconds per package, according to Adcock. "This is the last time a human will ever win," the Figure CEO predicted in his post on the competition. What lies ahead Not all the attention for the livestreamed robot demonstration has been uniformly positive. Some commenters pointed out moments when the robots seemed to mishandle packages and berated robots for occasionally dropping packages. Others have questioned whether the robots are truly operating autonomously, despite Figure's CEO insisting they operate independently, without any human teleoperators controlling their movements. The lack of independent verification on the ground for such robotic demonstrations often makes it difficult to confirm. There is also a history of some companies -- notably Tesla -- having relied on human teleoperators for many of their humanoid robot demonstrations. The Figure event may appear especially compelling because it's relatively rare for companies to present livestreamed endurance runs featuring humanoid robots. As this event demonstrates, such livestreams can seemingly convey greater transparency than short videos by allowing viewers to see robotic flaws and fumbles in real time. That may encourage viewers' belief in the robots' demonstrated capabilities, which in this case is limited to the Figure 03 robot's capability for handling packages in one specific warehouse-style setup. Even taking this particular demonstration at face value, what does this mean for Figure AI's broader vision? The company is one among many betting on AI-powered humanoid robots becoming general-purpose workers capable of performing various tasks normally done by humans. To gain widespread adoption, humanoid robots will need to prove as capable and cost-effective as either human workers or industrial robots with more specialized forms suited for specific tasks. This demonstration, focused on a repetitive task, does not show off such general-purpose capability involving a more diverse array of tasks or environments -- something that Figure has only hinted at in short video demonstrations so far. Still, the company has already raised nearly $2 billion from Silicon Valley investors and companies, including Microsoft, Nvidia, Intel, Amazon and OpenAI. Figure's biggest real-world test to date involved deploying its Figure 02 robots to the BMW Group Plant Spartanburg in South Carolina in 2025. Over an 11-month period, the previous version of Figure's humanoid robots focused on picking up sheet-metal parts from racks or bins and placing them on a welding fixture. The robots supposedly contributed to the production of 30,000 BMW X3 vehicles while working 10-hour shifts throughout a typical Monday to Friday workweek. "Figure 02 handled the precise removal and positioning of sheet metal parts for the welding process -- a task that is particularly demanding in terms of speed and accuracy while also being physically exhausting," according to a BMW Group press release. It remains to be seen whether BMW will also put the Figure 03 humanoid robot to work in more pilot deployments. In the February 2026 press release, the automaker said it was still evaluating future use cases for Figure's latest robot.

Figure AI humanoids sort 28,000 packages in 24-hour autonomous test

The system also includes automatic recovery mechanisms. Figure AI said if a robot gets stuck or encounters an unfamiliar situation, the AI system can trigger an autonomous reset and resume work without human intervention. "If the robot gets stuck or the AI policy goes out of distribution, Helix triggers an automatic reset," Adcock said. Figure AI further claimed the robots can independently leave the work floor for maintenance if software or hardware issues appear, while another robot automatically takes over operations to maintain uptime. "If a robot has a software or hardware issue, it autonomously leaves for maintenance and another robot takes over," Adcock wrote. The latest demonstration builds on Figure AI's earlier claims that its humanoid robots completed full eight-hour shifts autonomously using Helix-02. The company has also previously tested humanoid robots at BMW manufacturing facilities in South Carolina. Helix-02 is designed as a unified neural network combining vision, touch sensing, proprioception, and whole-body control. Unlike conventional industrial robots that separate movement and manipulation systems, Figure AI says its robots use a single AI model to handle walking, balancing, object handling, and coordination in dynamic environments.

Figure AI streamed humanoid robots sorting packages for 8 hours straight -- and not everyone is convinced it was fully real

* Figure AI F.03 robots have been sorting packages * An 8-hour livestream showed the bots working autonomously * The view count for the video is currently at 10 million Figure AI keeps setting new challenges for itself and its humanoid robots, and the latest test of these machines' dexterity and intelligence was just livestreamed on X: sorting packages, unsupervised, for a full 8-hour production-line shift. The feat was posted to X by Figure AI CEO Brett Adcock, and the three F.03 robots taking shifts here seem to have completed the job mostly as required. According to Adcock, the instructions were to detect package barcodes, pick up the packages, and then send them on their way down the conveyor belt, with the barcode face down. From the video I've seen -- which admittedly wasn't the full 8 hours -- the packages were all sorted correctly, bar a couple of exceptions. Adcock says the robots are now up to parity with human speed on this task, which is apparently 3 seconds per package, using integrated cameras and the on-board Helix-02 software. It's the full autonomy and lack of human supervision that Figure AI is making the most of here, and there are now plans to try the same challenge for a full 24 hours. These robots might soon be coming to a production line near you. Seeing is believing The livestream is certainly quite hypnotic to watch -- and it's also a hit, with 10 million views on the original video, prompting one Redditor to quip that the bots are "stealing jobs from warehouse workers AND streamers". There are several other entertaining reactions online to the activities of these robots. Some observers seem convinced that the machines are being operated remotely by humans, with one of the F.03 models apparently touching its head at one point -- a sign of a VR headset, if you think this is a scam. I'm not convinced by that argument, and Figure AI has no reason to lie about what's going on here. There are also reactions commenting on how much smarter and faster the robots are working compared to a previous demo along similar lines, evidence of progress on both the hardware and software sides. There are plenty of suggestions from commenters, too, including ways to improve the robot's design and efficiency, and doubts that these machines are anywhere near human speeds for package sorting at the moment. Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews, and opinion in your feeds.

Helix-02 humanoid robot handles full 8-hour factory work shifts

In a post on X, the California-based robotics startup wrote: "Watch a team of humanoid robots running a full 8-hr shift at human performance levels. This is fully autonomous running Helix-02." The announcement builds on Figure AI's latest Helix-02 system, a unified neural network that allows humanoid robots to walk, manipulate objects, balance, and coordinate movements continuously using onboard sensors. Unlike traditional industrial robots that rely on separate controllers for movement and manipulation, Helix-02 combines vision, touch, proprioception, and whole-body control into a single learning system designed for long-horizon tasks in dynamic environments. According to Figure AI, Helix-02 enables robots to complete multi-minute tasks without resets or human intervention. In one demonstration, a humanoid robot autonomously unloaded and reloaded a dishwasher in a full-sized kitchen over four continuous minutes.

Factory worker defeats humanoid robot in 10-hour package-sorting contest

An intern at U.S. robotics startup Figure AI narrowly beat one of the company's humanoid robots in a 10-hour package-sorting contest livestreamed online. In a post on X last week, Figure AI CEO Brett Adcock said the contest required visualization specialist intern Aime Gerard and humanoid robot Bob the Bot to sort as many packages as possible within 10 hours. The challenge followed California labor laws, meaning Gerard was given paid breaks and time to eat during the event. Adcock said the robot operated fully autonomously and was not remotely controlled by another human. Gerard led for most of the contest, though the robot briefly moved ahead after he stepped away for a bathroom break around the halfway point. By the end of the challenge, he had sorted 12,924 packages, 192 more than the humanoid robot. He averaged 2.79 seconds per package, while the robot averaged 2.83 seconds. Adcock later joked on X that Gerard's "left forearm is basically broken" and that his fingers were covered in blisters after the contest. "This is the last time a human will ever win," Adcock wrote. The livestream sparked debate online. One viewer wrote that the demonstration proved the robot was capable of doing "actual work," according to British news outlet Metro. Another commenter argued the humanoid still performed poorly, claiming it dropped four times more packages, struggled to flip square boxes and damaged more labels than the human worker. A separate user described the narrow loss as "a moral victory for the robots" and predicted humans would soon fall behind. Figure AI has also been livestreaming multiple humanoid robots sorting packages continuously at its headquarters since last Wednesday. The company aims to show potential customers that the robots can operate reliably for long periods, including 24-hour shifts, investor and board member Jesse Coors-Blankenship told Business Insider last week. Adcock said the livestream would continue until the robots fail. Ayanna Howard, dean of the College of Engineering at Ohio State University, said the robots' ability to work for long periods without failure was impressive, but added that the technology was not ready for large-scale deployment because of accuracy problems. Howard cited issues including packages being placed barcode-side up and boxes being knocked off conveyor belts. "We're a long way away from a fully autonomous humanoid in a logistics center."

Figure AI's Helix-02 humanoid robots is pulling full 8-hour factory shifts without human help

Figure AI has made one of the most aggressive claims recently regarding the capabilities of humanoid robots driven by the Helix-02 AI system. According to the company's statement, robots based on the Helix-02 AI system can independently operate for an entire eight-hour shift without any assistance from humans. Also read: OpenAI wants to build the UN of AI: Good luck with that The main difference between Helix-02 and robotic systems introduced in the past is architecture. Most traditional industrial robots have two separated controllers: one for motion and another for manipulation. However, the Helix-02 system integrates both controllers into one neural network and allows processing all the inputs simultaneously, including vision data from cameras located on the head and palms, tactile sensors on fingertips, and joint position data throughout the body. The advantages of Helix-02 were demonstrated in several demonstrations of autonomous humanoid robot operation. For example, one robot driven by the Helix-02 AI system autonomously unloaded and then loaded a dishwasher during four consecutive minutes without interruption and resetting. Another demonstration included two robots working autonomously and simultaneously within two minutes on rearranging a bedroom: hanging clothes, making a bed, emptying the waste bin, and arranging the furniture. Not only that but Figure AI also released System 0, an end-to-end whole body neural controller with over 1,000 hours of human motion data training. This reduces more than 109,000 lines of handwritten C++ code. This should not be taken lightly since handcrafted codes tend to be fragile, whereas a learned system generalises and can handle the chaos of real-life environments that scripted logic cannot do. Also read: LG's Sanjay Chitkara on AI making appliances smarter and building products for India Fine motor skills are also remarkable. Helix-02 robots are capable of unscrewing bottle caps, extracting pills from a pill organiser, pushing accurate amounts of syringes, and manipulating metallic parts from a bin based on tactile and in-hand vision. Furthermore, the robot works on top of Figure AI's previous implementation, which had been working in BMW's plant in Spartanburg, South Carolina, handling over 10 hours per day, moving over 90,000 parts per day and producing over 30,000 cars. The race for humanoid robots in factories is very much on the table now with other companies such as Tesla, Agility Robots, and Apptronik, amongst others, competing aggressively.

This robot can do 8 hour warehouse shift without any breaks, leaving internet talking about job security

People also noticed the robot was slower and rougher than human workers while handling boxes. American robotics company Figure has shared a new video showing one of its humanoid robots performing manual labour tasks inside a warehouse setting. The robot, called Helix-02, was seen sorting packages and placing them carefully onto a moving belt. Figure claimed that a team of such robots can complete a full eight-hour shift while matching human performance levels. The clip quickly caught attention online as it offered another glimpse into how AI-powered robots are slowly entering real workplace environments. While the robot successfully completed the task shown in the video, the footage also highlighted some clear differences between machine movement and human handling during physical labour jobs in warehouses and factories today. In the video shared by Figure on its official X handle, the Helix-02 robot can be seen lifting packages one by one and adjusting their position before placing them onto a rotating conveyor belt. The robot makes sure that every package faces the correct direction before pushing it forward for the next step in the process. The company presented the clip as an example of how humanoid robots can handle repetitive tasks that are usually performed by workers in warehouses and logistics centres. Figure also suggested that multiple robots working together could complete long shifts without breaks. Also read: OpenAI is offering two months of Codex access for free, but there is a catch Even though the robot successfully carried out the task, the video also revealed certain limitations. The speed of the robot appeared slower than what is normally expected from a human worker performing the same activity. Another apparent problem with the machine was the process of handling packages. Unlike people, the machine was not very gentle during the handling of boxes. It is quite possible that sometimes the packages were being roughly thrown to the floor, which could lead to damage to goods packed in them. In the same way, the way the machine pushed boxes toward the conveyor line also appeared rather rough. Also read: Is Sam Altman using OpenAI partnerships to grow his personal wealth? Court filings raise questions Although there are many problems that can be seen with this technology, it still indicates the fast development of AI robotics. Moreover, the machines like Helix-02 can be used in those fields that require repetitive manual labour which is life-threatening. Through updates in the future, some problems such as poor movement and slow speed will surely be solved.