Latest news with #humanoidrobots
CNET
8 hours ago
- Entertainment
- CNET
Highlights From World's First Humanoid Robot Kickboxing Tournament
China recently hosted a kickboxing tournament for humanoid robots and we've compiled all the highlights in one place. Six Unitree G1 humanoid robots and the humans operating them fought for the top spot in what was billed as "the world's first combat competition exclusively featuring humanoid robots." Each "fighter" was a Unitree G1 EDU model, the most advanced robot in Unitree's G1 lineup, armed with special combat training including straight punches, hooks, kicks, and knees. Unitree G1 robots slug it out for the top spot. Unitree Robotics These robots can be controlled by voice command, motion-sensing or remote control. Unitree says each control method has its pros and cons. Remote controls are being used to control the robots in this competition because they have lower latency than voice commands and are easier to learn. I got behind the controls of Unitree's G1 robot at CES and the company's Go2 Pro robot at CNET HQ. I found the remote controls very intuitive and similar in feel to the video game controllers I grew up with. "AI Strategist" in black fought "Power Core Guardian" in green for the top prize. Unitree Robotics After four matches of punishing robot-on-robot carnage, including at least one spectacular knockout, the robot in black, nicknamed "AI Strategist," took home the top prize. Check out the video in this article to see all the gnarliest highlights.
Forbes
20 hours ago
- Entertainment
- Forbes
Watch The World's First Humanoid Robot Kickboxing Match
Unitree G1 robots in a kickboxing competition in China. 52% of U.S. workers are worried about robots replacing them on the jobsite, according to a recent Pew Research finding. Now we might have to add UFC fighters to those numbers, since China's Unitree Robotics staged the world's first-ever humanoid robot kickboxing match. The robots, honestly, aren't yet amazing at punching and kicking each other, and sometimes fall off-balance while attacking or defending themselves. But, like the humanoid robotic half-marathon held in Beijing in mid-April, it's a start. And with the pace of innovation in humanoid robot development, it's going to get much better. The kickboxing match featured four Unitree G1 robots, which are relatively small for humanoid robots at just over four feet tall and under 80 pounds in weight. Their punches generally lacked power because the G1 robots are fairly slow, meaning they essentially pushed each other with their boxing glove covered hands rather than actually snapping a punch with pace into their opponent's heads. Kicks were similar, and some kneeing attacks just hit thin air as their robotic opponent spun out of the way. However, the G1s featured excellent return-to-standing ability after slipping or being knocked down, even when getting tangled up in the boxing ring ropes. For now, the humanoid robot kickboxing competition is something interesting, but not for its combat entertainment value. There isn't a huge amount of that, but the fact that the robots are fighting at all is astonishing. And it's somewhat promising. I love combat sports, but it's a guilty pleasure. I've mostly stopped watching mixed martial arts competitions, because it's just so destructive to human health and wellbeing: two fighters who have trained themselves to the peak of human strength, flexibility, and capability proceed to utterly destroy each other, with potentially long-term health and wellbeing impacts, particularly to their brains. Humanoid robot combat sports might offer the interest factor without the inevitable human carnage that results. The fight was broadcast on Chinese state television, Asia Times says. Fight training was handled by AI: 'It is not easy to teach robots different movements,' Wang Qixin, a director at Unitree, told Chinese Central Television. 'We used artificial intelligence (AI) technology to train them.' Innovation in humanoid robots is ramping up significantly, Apptronik CEO Jeff Cardenas told me in a recent interview, and AI is a major driver of that. 'I think we're on this exponential curve,' he says. 'I think the simple way of explaining it is we're moving from pre-programming robots and modeling the world to where the robots can now learn [themselves].' Apparently, that's not just how to move boxes, tighten screws, or mop floors. It's also how to throw a punch and deliver a kick.
Forbes
21 hours ago
- Business
- Forbes
Humanoid Robots Is The ‘Space Race Of Our Time,' Says Apptronik CEO Jeff Cardenas
The struggle to be first in humanoid robots is the space race of our time, says Apptronik CEO Jeff Cardenas, who also says that major upgrades are coming to Apptronik's appropriately-named Apollo robot in 2025. 'It is just amazing to me to hear that there's a hundred companies working on humanoid robots,' Cardenas told me recently on the TechFirst podcast. 'Investors that I talked to two years ago that said humanoids don't make sense. They didn't wanna pay attention to hardware. Now they have a humanoid thesis and hardware's the name of the game.' Apptronik closed a massive $403 million funding round early this year with blue-chip investors including tech titans like Google, massive automotive brand Mercedes-Benz, and venture heavyweights such as B Capital and Capital Factory. The company has also partnered with $30 billion global manufacturing giant Jabil to ramp up production capabilities and achieve mass scalability. 'One of the things that we realized was that if we really are going to ramp to tens of thousands, hundreds of thousands of units, we need to really learn about how to do manufacturing right and do it at scale," Cardenas told me. There are two main challenges now in humanoid robots. One is the breakneck race to bring the best, fastest, smartest, most capable and most adaptable humanoid robot to market, and that's dominated by companies like Figure, Tesla, Agility Robotics, Apptronik itself, and about 20 other companies in humanoid robotics. A European manufacturer, Neura Robotics, says it will ship a 'best in world' humanoid robot this summer. The second is to manufacture at scale to get the price affordable. Bank of America says that is likely to happen by 2028, when humanoid robots begin a mass adoption trend for commercial use. Humanoid robots, Bank of America says, can replace 20% of the world's industrial sector jobs: roles that employ perhaps 800 million people today. Service sectors and the home sector will come next, with a project penetration rate of .7 humanoid robots per household, and this phase is projected to kick off in 2034. Figure alone plans to ship 100,000 humanoid robots in the next four years: perhaps optimistic, but intriguing nevertheless. One thing we can say about Apptronik is that it has the capital and the manufacturing relationships to boost its odds of winning. Powered by recent AI breakthroughs, Apptronik is shifting beyond simple tasks like box-moving to advanced dexterous tasks. Cardenas emphasizes a human-centered design philosophy, aiming to create robots people want to be around, and expects meaningful industrial deployment in 2026, with additional use cases in healthcare, hospitality and elder care to follow 3–5 years later. Apollo improvements we'll see this year include upgraded dexterity, modular batteries for 24/7 uptime, and a design robust enough for factory floors or fulfillment centers. While clearly there's the fear of human replacement and job loss, Cardenas sees humanoid robots as helper, not usurpers. 'These are tools to augment human productivity and capability,' he told me. 'I think that's particularly important for humanoid robots is that they're thought of and designed to be human helpers versus something that replaces humans.'
Yahoo
a day ago
- Automotive
- Yahoo
Elon Musk Thinks Tesla Will Be the World's Most Valuable Company, but This Huge Problem Could Send Its Stock Plunging by 70% Instead
Self-driving cars and humanoid robots are key to Musk's bullish vision for Tesla, but the company has to overcome major challenges first. Tesla's electric vehicle sales are plummeting, and continued weakness could lead to massive losses for the stock. These 10 stocks could mint the next wave of millionaires › Tesla (NASDAQ: TSLA) stock hit a new record high in Dec. 2024 on the back of President Trump's election win. Investors speculated that Trump's focus on deregulation could speed up the company's ability to commercialize its autonomous robotaxi and humanoid robot opportunities. In fact, CEO Elon Musk believes those two businesses will eventually make Tesla the world's most valuable company one day -- potentially even more valuable than the next five companies combined. Today, those five companies are Microsoft, Apple, Nvidia, Amazon, and Alphabet, and they have a combined market capitalization of $13.6 trillion. But there's a problem: Tesla is worth only $1.2 trillion as of this writing, and most of its revenue still comes from selling electric vehicles (EVs), which are seeing plummeting demand. That's bad news for Musk's ambitious forecast in the near term, and here's why it could lead to a collapse of over 70% for Tesla stock instead. Back in 2023, Musk told investors Tesla could grow its EV production 50% annually for the foreseeable future. Deliveries hit a record high of 1.81 million vehicles that year, a year-over-year increase of 38%. But in 2024, deliveries shrank by 1% to 1.79 million. Simply put, Tesla can't grow production 50% per year if sales aren't keeping up. Unfortunately, the situation appears to be getting worse: Deliveries plunged even further in the first quarter of 2025, this time by 13%. And there could be an even sharper decline in the second quarter if the available data from April is accurate. In the United Kingdom, new Tesla EV registrations fell 62% year over year in April. There were also sharp declines across Europe -- registrations collapsed 81% in Sweden, 74% in the Netherlands, 67% in Denmark, and 59% in France. Analyst estimates indicate Tesla might deliver as little as 350,000 cars during the second quarter, or a year-over-year drop of more than 20%. The problem seems to be specific to Tesla because EV sales overall soared 28% across Europe in April. Chinese automaker BYD saw a staggering 359% increase in sales across the region, and it sold more cars than Tesla for the first time ever during the month. BYD is known for its ultra-affordable models, like the Seagull, which sells for under $10,000 in China. Tesla simply can't compete at that price point. Musk's political affiliations might also be damaging the brand, mainly because of his work within the Trump administration's Department of Government Efficiency. Musk helped slash thousands of federal jobs and shuttered entire agencies. According to the most recent CNBC All-America Economic survey, around half of Americans now have a negative view of Musk, while only 36% view him positively. That public opinion flowed through to Tesla with 47% of respondents expressing a negative view of the company. Last October, Tesla unveiled its long-awaited Cybercab robotaxi. It has no pedals or steering wheel because it's designed to run entirely on the company's Full Self-Driving (FSD) software. Musk wants to build a ride-hailing network in which Cybercabs can transport passengers and earn revenue for Tesla around the clock. But the unsupervised version of Tesla's FSD isn't approved for use on public roads yet. Musk says the company could offer paid autonomous rides starting in June of this year, but he doesn't expect true scale to be achieved until the second half of 2026. That places Tesla significantly behind the likes of Waymo, which is already completing over 250,000 paid autonomous trips every single week across Los Angeles, San Francisco, Phoenix, and Austin. Catching up to Waymo won't be easy because it has partnered with Uber Technologies, which operates the world's largest ride-hailing network with 170 million monthly users. It could take Tesla years to achieve that level of scale, especially if it builds its own mobility network from scratch. But Musk is eyeing another major opportunity: the Optimus humanoid robot. He says it could generate a staggering $10 trillion in revenue over the long term. Tesla will reportedly manufacture thousands of them this year to use inside its factories, but Musk predicts the company could be producing millions of them annually by 2029 or 2030. Optimus could eventually be used for the dangerous jobs and repetitive tasks that humans don't want to do, so there is an enormous addressable market across the business and consumer sectors. While Musk claims humanoid robots will outnumber humans by 2040, keep in mind the CEO is known for bold predictions that don't always come to fruition. If products like the Cybercab, FSD, and Optimus are a success, Tesla could certainly become the world's most valuable company. But it's facing a serious problem right now because its stock is trading at an eyewatering price-to-earnings (P/E) ratio of 186.5. For some perspective, Nvidia, Microsoft, Amazon, Apple, and Alphabet trade at an average P/E ratio of just 32.2: Tesla's earnings per share (EPS) plunged 71% during the first quarter of 2025, whereas each of the other five companies generated EPS growth in their most recent quarters. Tesla's premium valuation makes even less sense from that angle because investors typically assign a higher P/E ratio to companies that are growing quickly. Tesla stock would have to plummet by 76% just for its P/E ratio to trade in line with Nvidia's, and that is a legitimate risk. Despite the long-term potential of products like the Cybercab and Optimus, 72% of Tesla's revenue still comes from selling EVs, and the pace at which sales are shrinking will almost certainly drive an ongoing collapse in the company's EPS for the foreseeable future. Remember, based on Musk's optimistic predictions, it could be years before the Cybercab and Optimus businesses are generating enough revenue to offset the weakness in Tesla's EV sales. As a result, there is little hope the company will grow to exceed the combined value of Nvidia, Microsoft, Amazon, Apple, and Alphabet anytime soon. I believe the stock is destined for a sharp decline instead, bringing its P/E ratio in line with those of its big-tech peers. Ever feel like you missed the boat in buying the most successful stocks? Then you'll want to hear this. On rare occasions, our expert team of analysts issues a 'Double Down' stock recommendation for companies that they think are about to pop. If you're worried you've already missed your chance to invest, now is the best time to buy before it's too late. And the numbers speak for themselves: Nvidia: if you invested $1,000 when we doubled down in 2009, you'd have $349,597!* Apple: if you invested $1,000 when we doubled down in 2008, you'd have $38,048!* Netflix: if you invested $1,000 when we doubled down in 2004, you'd have $651,761!* Right now, we're issuing 'Double Down' alerts for three incredible companies, available when you join , and there may not be another chance like this anytime soon.*Stock Advisor returns as of May 19, 2025 John Mackey, former CEO of Whole Foods Market, an Amazon subsidiary, is a member of The Motley Fool's board of directors. Suzanne Frey, an executive at Alphabet, is a member of The Motley Fool's board of directors. Anthony Di Pizio has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Alphabet, Amazon, Apple, Microsoft, Nvidia, Tesla, and Uber Technologies. The Motley Fool recommends BYD Company and recommends the following options: long January 2026 $395 calls on Microsoft and short January 2026 $405 calls on Microsoft. The Motley Fool has a disclosure policy. Elon Musk Thinks Tesla Will Be the World's Most Valuable Company, but This Huge Problem Could Send Its Stock Plunging by 70% Instead was originally published by The Motley Fool
Bloomberg
2 days ago
- Business
- Bloomberg
China's Startups Race to Dominate the Coming AI Robot Boom
A few weeks ago, a Chinese startup called EngineAI gave an unusual demonstration of how today's robots learn. A female instructor with long blonde hair showed off a series of dance steps and then encouraged the company's humanoid robot to imitate her moves. 'Five, six, seven, eight,' she said. 'Let's go! Keep the flow.' Using computer vision and machine learning algorithms, the machine watched and listened — and then mimicked a series of steps from her dance routine. Over a few days, EngineAI's robot mastered the Axe Gang dance from the 2004 movie Kung Fu Hustle. A bit silly perhaps, but the dance represents a serious step for China's tech industry. After years of American companies like Boston Dynamics Inc. leading the development of humanoid — or people-shaped — machines, Chinese upstarts are now pushing the boundaries of innovation. In April, Beijing hosted the world's first robot half-marathon, where the city's X-Humanoid bested a field of 20 other bipedal machines. In May, Unitree Robotics, one of the country's prominent players, showcased its machines in what was billed as the first robot kickboxing tournament. While these events don't always go smoothly — 15 of 21 robots failed to complete the Beijing race — the point is progress, not perfection. China, which already has a higher density of robots per human on its factory floors than the likes of the US and Japan, is preparing humanoids to move into increasingly complex roles. EngineAI, Unitree and their competitors have started trials for everything from sorting garbage and delivering medicines in nursing homes to patrolling the streets alongside police officers and guiding tours through museums. The bots are quietly being tested for military combat, according to local media reports. Zhao Tongyang, founder and chief executive officer of EngineAI, figures 50 or 60 companies in China are working on humanoid development now, benefiting from the country's manufacturing expertise and robust government support. They're harnessing artificial intelligence models so the machines can learn how to handle new tasks on their own, without laborious programming for each situation. The company says it has orders for hundreds of its humanoids. 'China has many players involved and there are some very good talents among them,' said the 43-year-old Zhao during an interview over Zoom. 'I hope China will be first when it comes to robots.' The country's startups have caught the attention of Elon Musk, whose Tesla Inc. has set its sights on the humanoid market. On an April conference call, the billionaire said he thinks his Optimus robots lead the industry in performance, but China may end up dominating the field. 'I'm a little concerned that on the leaderboard, ranks 2 through 10 will be Chinese companies,' he said. Leadership in this field matters because humanoids appear poised to move beyond the realms of sci-fi and curiosity. Citigroup Inc. recently projected the market for the machines and related services will surge to $7 trillion by 2050 when the world could be populated by 648 million human-like bots. Beyond the monetary rewards are mind-rattling strategic implications for governments and countries. Whoever leads development of these humanoids could potentially command an army of indefatigable workers, caregivers and soldiers, redefining economic and political strength. Musk made the case in mid-May that the global economy could grow to 10 times its current size with the addition of robot labor. 'It unlocks an immense amount of economic potential,' Musk said during an appearance in Saudi Arabia. 'We're headed to a radically different world.' China's strength in this field isn't an accident. President Xi Jinping and the Communist Party sketched out blueprints for developing strategic technologies — including robotics — more than a decade ago. Local officials then offered financial incentives and support to entrepreneurs to reach Beijing's policy goals. In EngineAI's case, the Shenzhen government helped connect Zhao with funding and other resources. 'Although it is not the government that directly invests in us, at least the government's money is used to guide them to this industry and guide funds in this direction,' the founder said. 'I think this is a great thing.' China announced earlier this year it would invest 1 trillion yuan ($138 billion) in robotics and high tech in the next two decades, far more than the US or Europe. The US has formidable players in robotics — including Boston Dynamics, Agility Robotics, Figure AI and Tesla — and a long track record of technology breakthroughs. Yet some scholars warn that Beijing's approach may give China the edge in developing strategically important, capital-intensive sectors, like it has already done with electric vehicles and solar panels. 'The Chinese model of state-run capitalism might actually be better equipped,' said Julian Mueller-Kaler, director of the Strategic Foresight Hub at the Washington D.C.-based think tank, Stimson Center. 'Digital and tech advancements are geopolitical issues of the highest order.' While it's still possible the humanoid market never takes off, China is making an audacious bet that it will. The country is on track to produce more than 10,000 humanoid robots this year, or more than half of the machines globally, according to an April study from the China think tank Leaderobot and other institutions. 'China is winning the humanoids war, I have no doubt,' said Henrik I. Christensen, director of the Contextual Robotics Institute at the University of California San Diego. Zhao's career mirrors the country's embrace of robotics. After majoring in automation at university, he got into the field eight years ago, motivated not so much by Beijing's strategic priorities as the prospects for innovation. 'I believed that humanoid robots would definitely change the way of life for human beings,' he said. His timing was fortuitous. A few years earlier, Beijing had unveiled an ambitious program called Made in China 2025, which set goals for technological accomplishments over the next decade. The Communist Party's drive to develop humanoid technology stemmed from a looming labor crunch. The working-age population is expected to shrink by about 22% through 2050, according to the Lowy Institute in Sydney. The squeeze is particularly acute in manufacturing, with a projected shortfall of 30 million workers in 10 sectors by the end of this year, according to a report from government agencies, including the Ministry of Industry and Information Technology. This gap is aggravated by the younger generation's reluctance to take the kind of blue-collar factory jobs their parents embraced. China's strategy is to integrate people-like robots across sectors like manufacturing, healthcare and hospitality. Factory floors in China already boast a robot density that eclipses Germany and Japan, with levels doubling in just four years. It had 470 robots per 10,000 employees in 2023, well above the US at 295 robots for the same number of workers, according to the International Federation of Robots. Artificial intelligence models are helping humanoid robots become smarter, more adaptable machines. Startups like EngineAI and Unitree are refining their AI applications, allowing robots to see and recognize objects, plan movements, coordinate with each other and teach themselves how to adapt to new situations. 'Previously, robots were very stupid, right?' said Chang Lin, the co-founder and CEO of Leju Robotics, another ambitious startup. Today, the evolution of large language models, like those powering ChatGPT and DeepSeek, is bestowing intelligence on machines so they can learn tasks from their owners, he said. You could, for example, train one of the company's Kuavo robots how you want it to clean the floors or care for the flowers in your house — and then it would take over. 'It'll be easy,' he said. 'Form a data set for watering flowers, it will naturally water the flowers.' This kind of progress has China's robots making their way into the real world. UBTech Robotics Corp., based in Shenzhen, has deployed its 5-foot-6-inch Walker S1 to help assemble iPhones for Apple partner Foxconn Technology Group. More than 500 of its machines work in the auto factories of BYD Co., Geely Automobile Holdings Ltd. and FAW-Volkswagen, lifting boxes, sorting components and testing instruments. At the FAW-Volkswagen plant, they detect refrigerant leaks in air-conditioning systems, a task that could hurt the lungs of humans. Leju, also based in Shenzhen, is pushing the use of humanoids to sort and deliver medicines at elder-care facilities. At least a hundred of its machines have been delivered to carmakers including BAIC Motor Corp. and Nio Inc., exhibition halls in China and nursing homes in Suzhou. Colleges have also bought Leju robots for doing research. Susanne Bieller, general secretary of the Frankfurt-based International Federation of Robotics, sees China emerging as a leading player in humanoids as the country's startups work with its manufacturers to bring down costs and make machines more affordable. 'Within the next 5 to 10 years, we can expect it to be more widely adopted in industry,' she said. EngineAI spent years improving its robots' joints, making them lighter and smaller, while whittling away at expenses. In one case, Zhao and his 40-person engineering team redesigned a power joint that cost 20,000 to 30,000 yuan so they could produce it themselves for one-tenth the cost. Zhao also grew obsessed with figuring out how to create robots that could walk with a natural, human-like gait. He told staff early machines that take slow, awkward steps should be thrown in the trash. EngineAI built humanoids with refined leg joints and then integrated machine learning with neural networks so the machines could learn to walk, much like a real toddler. 'Collect a lot of very beautiful data from the human body, combine these data and let it walk,' he said. Still, even the most elegant humanoids won't have a future unless they provide value. People-like machines captured the popular imagination at least as far back as Isaac Asimov's writings in the 1950s, yet they've remained largely a novelty. Boston Dynamics has impressed tech geeks since its founding in 1992, but it's never built much of a business. Google and SoftBank Group Corp. each bought the startup and then sold it again without commercial success; it's now owned by Hyundai Motor Co. In Beijing, the humanoid half marathon meant to showcase the machines' capabilities instead demonstrated their limitations. One robot participant fell at the starting line. Another had its head fall off and roll on the ground. In the kickboxing tournament this month, the Unitree robots often lashed out at thin air or toppled over on their own, even though they were controlled by humans ringside. The vibe was less Terminator, more Rock 'Em Sock 'Em Robots. Humanoids 'just don't make economic sense for most people and companies for the foreseeable future,' said Romain Moulin, CEO of the French startup Exotec, which makes box-like robots for warehouses that he thinks are more utilitarian. Nevertheless, the number of believers is growing. In its 83-page report, Citigroup tapped 10 of its own staff and seven contributors to explain why it sees robots as a yawning opportunity. It describes such machines as 'physical AI' and predicts they will be used for autonomous driving, cleaning and deliveries. It forecasts that humanoids will be the fastest-growing segment of the robot market, with the flexibility to work in health care, factories, home cleaning, deliveries, supermarkets and hospitality. 'A big part of the thesis for humanoids is simple – we have engineered our manmade world to work for humans so humanoids can fit straight in without significant infrastructure changes,' the authors wrote. Musk has become a leading advocate. He said in April he expects to have thousands of Optimus robots working in Tesla factories by the end of the year, and he is confident of getting to a million units a year in less than five years. The robots aren't on sale for others yet, but they're expected to cost $20,000 to $30,000 as they hit the market in 2027. The economics of such machines will then become clearer, Citi contends. If Musk's prediction is accurate that humanoids cost about $25,000, the analysts calculate they would pay for themselves in 36 weeks, if compared with the lowest US minimum wage of $7.25. The payback equations are more compelling in the states and jobs where wages are higher. 'The argument for humanoids is not just economic. Some jobs are dangerous or undesirable for humans and are better suited for robots. In other cases, labour shortages exist and can fill these roles. Robots also don't leave or take holidays,' they wrote. Chang, the Leju Robotics CEO, said the growing number of humanoid startups in China is driving intense competition. Scores of companies have to race their rivals to integrate AI capabilities and physical innovations. 'The more people enter the industry, the better the industry will be,' he said. 'So what we need to do is to try our best to stay ahead.'— Saritha Rai and Jessica Sui
