How India Is Training the Robots of the Future

0:00Some people treat ChatGPT like some kind of smart search engine, and some use it to get work done. ChatGPT Work is a new way of working in ChatGPT that can take action across your apps and files, stay with a project for hours if needed, and turn a goal into finished work. It's designed to help you move from a chaotic starting point to a reviewable first version. So all the source materials, briefs, and scattered information that you have to grind through to turn into something useful can just become something useful. Put ChatGPT to work on your most ambitious ideas and projects. Get started at ChatGPT.com by selecting Work Mode, available on Plus and Pro plans.

0:42This is Robert Smith from Business History. If you're listening to this, there's a good chance you're a small business owner. And like every small business owner, you started with a dream to do what you love and watch it grow. What you probably didn't dream about? Keeping up with cyber threats. That's where MasterCard can help with access to tools that help identify cyber threats to better protect your business. Building your dream business? Priceless. For cybersecurity in a changing world, there's MasterCard. Learn more at MasterCard.com slash small business. As industries evolve faster than ever, companies need an environment that accelerates strategic growth, and Michigan delivers on that promise.

1:19From emerging startups to global enterprises, Michigan offers what executives value most, a resilient, innovative ecosystem, diverse communities that attract top talent, and a quality of life that supports work-life balance. With our unified Team Michigan approach, businesses scale faster and compete at the highest level. Michigan. Pure opportunity. Seize your opportunity at michiganbusiness.org. Bloomberg Audio Studios. Podcasts. Radio. News. On top of a mountain of discarded plastic, a 43-year-old woman named Sunita Rattour is hard at work. She's in a bright yellow sari with green glass bangles sifting through piles of litter.

2:05Plastic waste which she'll sort and clean for recycling. Sunita is based in New Delhi, but scenes like this one take place every day all over India. On construction sites, in homes and recycling plants where manual labour is done. Nothing out of the ordinary, except for one thing. Suddenly you saw this high-tech contraption on her forehead, an iPhone. An iPhone strapped to her head, recording every move she makes. Saritha Rai covers tech in India for Bloomberg. She met Sunita at a recycling plant earlier this year. Sunita has spent years sorting and cleaning discarded plastic for about$200 a month.

2:50Usually, the factories where she works are lively places. There's a lot of banter. There's a lot of conversation. There's a lot of silliness that goes on. But in this, it was deathly quiet because all these women were not allowed to talk to each other. Eight hours a day in total silence. But that's not all. She was hunched forward constantly because she was not allowed to capture any other part of her body except her wrist. It's exactly what a humanoid robot's camera would see if it looked down. That was the motion she was asked to capture. Sunita doesn't know what the footage is for. All she knows is that this phone recording earns her extra money, money she uses to send her four kids to school.

3:38But the recordings she and her fellow workers make are critical to the future of AI and robotics. They'll be fed into AI systems to teach humanoid robots how to move and navigate the physical world, like us, one frame at a time. The goal really is to capture the micro gestures, the really tiny hand movements that, you know, robots need to learn to become better at everyday tasks. The world has started to embrace chatbots and agentic AI systems like Claude or ChatGPT, which rely on cloud computing or local hardware to run. You give an AI agent a voice, text prompt or task, it gives you an answer.

4:24And for years, the robotics industry has tried to create humanoid devices, AI that moves in the physical world the same way that we do. When it's gone well, it's been dazzling. OK. We have a new world record. At the World Humanoid Robot Games in Beijing last month, essentially an Olympics for machines, a Chinese robot shattered the 100-metre sprint record. Its time? 8.64 seconds. Nearly a full second faster than Usain Bolt, the fastest man on Earth. But step outside that sterile, controlled track. And it's a completely different story. When robots come into the real world, there's invariably a blooper reel awaiting the texture of real life, the unevenness of perhaps the terrain, the spatial intelligence that's required for these robots.

5:25All of these things, you know, make it really difficult for robots to be trained in the lab. They have to be trained in the real world with real world data, which is where, you know, all of these data collectors come in. Citigroup projects the market for human-like robots could reach$7 trillion by 2050. And to learn how to perform tasks like a human, robots need huge amounts of so-called egocentric data, footage captured directly through the eyes of a person. China has been the biggest generator of that data, but it's keeping it within its borders. So the demand has shifted to the world's most populous country.

6:10So India really has become the epicenter of data collection in the past few months. Tens of thousands of workers like Sunita, shoe factory workers, denim factory workers, construction workers, all kinds of people have been recruited to record these first person camera footage, what is called egocentric footage. It was a sort of unsettling moment for me when I talked to these workers because most of them were recording themselves into their own job extinction.

6:48Welcome to the Big Take Asia from Bloomberg News. I'm Rebecca Chung-Wilkins, in for one heart. Every week, we take you inside some of the world's biggest and most powerful economies and the markets tycoons and businesses that drive this ever-shifting region. Today on the show, why the road to the next AI breakthrough runs through India and what this race for human data means for the US-China tech rivalry.

7:21Humanoid robots and online chatbots both use AI. But Bloomberg's Saritha Rai says teaching a large language model to write an essay is completely different from teaching machine to cook a meal. Real-world humanoid robotics is a completely different ballgame. This requires real world data, actual human motion and human movements that have to be learned by the models to teach the robots. There is no availability of any of this data online, you know, on the Internet. You might think, wait, the Internet has plenty of data. All those videos on YouTube or TikTok, all those humans doing random things.

8:03But... It's really raw data and it's not first-person egocentric data. It is not the data that a humanoid robot's camera would see when it's looking down or looking forward. That kind of data has to be collected in a highly standardised way. Faces can't appear in the footage and conversations aren't allowed. Equally important, the data has to come from as many different humans as possible. Each person has their own way of doing things, like how you perhaps open a can could be completely different from how I open a can. So each person's data is valuable and multiple labs are looking for what is called proprietary data.

8:48Saritha says India is an ideal place to gather this kind of data. The country has an enormous population, labour is cheap and workers still do a lot of manufacturing by hand. The diversity, whether it's the shoe factories, the denim factories, whether it's the, you know, warehouses of different kinds, whether it's the households, fishing villages or the farms or construction sites or the mines. It's incredible the diversity of tasks that can be recorded in India. No other country could offer this kind of a diversity of manual labor, not even China, I would say, because a lot of these tasks, particularly in factories and warehouses, are already automated in some way or the other in China.

9:33And what about the cost? Can you break down what kind of cost you have if you're buying this data in India versus somewhere else, say Singapore or even the US? I haven't looked at the costs in Singapore, but I can definitely talk about the costs in India versus the US. For instance, the raw first-person footage of the ecocentric data that India provides could come for about$7 an hour. But if it's cleaned and annotated, it could go for as much as$30 an hour. That's a huge value addition right there. And the same data, you know, could go from two to three to five times more expensive in the US when it's collected in the US.

10:19Tesla, for instance, pays$48 an hour in Palo Alto for motion capture data collected in-house. The factory floor is just the starting point for these egocentric videos. Once the footage is captured, it's sent to local firms that clean it up, check the quality and label the data so it can be used to train AI models. Demand for this work has surged, helping fuel a boom in India's AI services sector, with a growing number of companies built around processing and supplying data to tech firms. Who are these startups working with? What's the sort of end point of all of this proprietary data? Now, none of these companies were able to tell me who exactly was the end customer.

11:05Not that they didn't know, but the NDAs were so strict and so binding that none of them were allowed to reveal. But enough hints were dropped that these were all like major American and global companies, including Google and Tesla, Nvidia and Meta. You know, Nvidia, for instance, is now getting into physical AI and embodied AI models. So those kinds of companies would be the ultimate buyers of these cleaned, annotated data sets that then can be used as training material for building these physical intelligence models.

11:44While India has become the world's outsourcing hub for egocentric data, China is playing a very different game. After the break, why Beijing insists this data stays inside its borders.

12:01Some people treat ChatGPT like some kind of smart search engine, and some use it to get work done. ChatGPT Work is a new way of working in ChatGPT that can take action across your apps and files, stay with a project for hours if needed, and turn a goal into finished work. It's designed to help you move from a chaotic starting point to a reviewable first version. So all the source materials, briefs, and scattered information that you have to grind through to turn into something useful can just become something useful. Put ChatGPT to work on your most ambitious ideas and projects. Get started at ChatGPT.com by selecting Work Mode, available on Plus and Pro plans.

12:43This is Alexis Christophorus for Bloomberg Surveillance. AI is everywhere. Outcomes are not. Not because AI doesn't work, but because AI hasn't reached the workflows yet. The campaigns, the launches, the quarterly planning, those workflows are still run by humans alone. AI is making individuals faster, but it's not making businesses more productive. That's the gap Asana is built to close. Asana is the operating system for human agent teams, your easy button for AI productivity across every team. Ready-to-go AI teammates pre-built for marketing, ops, and IT. No prompt engineering, no setup. They show up where the work is happening, already onboarded in your workflows, ready to deliver.

13:26And the more they work with your team, the smarter they get. With Asana, your whole company can work on the same plan towards the same goal, whether you're a team of 10 or 10 ,000. Asana, where humans and agents workflow together. Try it at asana.com. That's A-S-A-N-A dot com. This is Robert Smith from Business History. If you're listening to this, there's a good chance you're a small business owner. And like every small business owner, you started with a dream to do what you love and watch it grow. What you probably didn't dream about? Keeping up with cyber threats. That's where MasterCard can help with access to tools that help identify cyber threats to better protect your business.

14:02Building your dream business? Priceless. For cybersecurity in a changing world, there's MasterCard. Learn more at MasterCard.com slash small business.

14:19Just last month, Beijing hosted the World Robot Conference. The event drew hundreds of exhibitors and showcased more than 3 ,000 products. Visitors saw robots making meals, sorting packages, and even facing off in ping-pong and boxing matches.

14:39The display wasn't just for show. China is rapidly becoming the epicentre of the humanoid robot industry. Chinese companies now account for more than 97 % of global humanoid robot shipments. Hundreds of companies are racing to move these machines from the showroom into factories or warehouses and everyday service jobs. Saritha says China keeps a tight grip on the data it gathers to develop these robots. Even though China has a massive manufacturing base and a humongous number of situations that can be recorded and used to train dropouts, there is a real fear that this data going out of the country would dim the manufacturing competitive advantage of the country.

15:28So there is stringent, stringent guidelines on how companies can use the data. And only local domestic Chinese operators are allowed to collect. There are guidelines on where they can collect, how they can collect. Can you just explain a little bit more? Why is there anxiety among Chinese policymakers to keep that data within China? The US and China have been running really hard. And as far as the large language models go, the U.S. has been clearly outrunning and outfacing China. The most advanced frontier models have always come out of the U.S. labs, whether it's OpenAI or Anthropik or one of those companies.

16:09The Chinese have always followed. Their models are just behind now, but not quite there. But on the other hand, in physical AI, which is robotics and drones and autonomous driving, China feels it has a real advantage. particularly because China has been making things physically for decades for the entire world. China feels that that manufacturing advantage is what is going to keep it ahead in physical AI. So as much as possible, China wants to keep the data within its borders that is being recorded. These will be only used to train Chinese physical AI models and will not be available to the world.

16:48The race between the US and China to dominate robotics is accelerating. In 2025, China installed nearly 300 ,000 industrial robots, roughly eight times the number deployed in the US, according to Robotics Center. Sarifa says China has some significant advantages, but it's still far too early to know who will ultimately come out ahead. For all the excitement, humanoid robots still aren't ready for widespread use. The robotics industry is still grappling with questions of cost, reliability and real-world utility. And many of the demonstrations on display in Beijing last month were carefully stage-managed.

17:31Still, at the conference, Unitree Robotics founder Wang Xingxing predicted the industry could hit its chat GBT moment within a decade. So I hope that, whether it takes as little as two or three years or as long as five or ten, we will reach a point where a robot can be brought into any unfamiliar environment, such as your home, and successfully perform around 80 % of the required tasks. I believe that would mark the chat GPT moment. So no robot is going to come and do my laundry for me anytime soon. I would not hold my breath right now for robots to cook clean and do my laundry for me yet. But the moment that the GPD models started becoming better and better enough for us to use the chatbot, chat GPD, and ask it anything, happened within a few years.

18:24And certainly when they become available and make economic sense in a country like India, where, you know, there is this whole layer of should we use robots or should we give humans work? This is a difficult question for policymakers around the world. How do you embrace technologies that boost productivity without displacing the very workers who helped build them? For millions of workers in India, that tension is especially acute. Saritha says there's a deep irony at play for people like Sunita, who are helping train the next generation of humanoid robots. The workers that are helping build the technology will actually make their labor less valuable eventually, particularly in a country like India, where there's already joblessness, there's already unemployment among Indian youth.

19:21and you just saw a whole uprising called the Cockroach Janata Party around this very core subject of education and joblessness, this could become a real critical theme in the coming years. How do workers like Sunita react when you tell them that they are potentially training robots that could put them out of a job. My mind keeps racing back to the moment I met her, you know, in that hot day in Delhi not too long ago. And she looked really happy to be doing what she was doing.

20:10For people like Sunita, where opportunities are very few, And suddenly their 20 ,000 rupees a month job becomes, you know, maybe a 60 ,000 rupees a month job. Suddenly with no extra effort other than the discomfort, of course, the great discomfort of wearing the camera all day in that heat. And, you know, craning their neck and hunching and, you know, recording themselves. But that is a real incentive. And Sunita, for one, was extremely happy that she could finally not need debt to send her four kids to school as the new school year was just coming up and there was school fees to pay, school uniforms to buy, books to buy.

20:56And she was happy that the work she was doing would help her comfortably navigate through those financial demands.

21:12This is The Big Take Asia from Bloomberg News. I'm Rebecca Chung-Wilkins. To get more from The Big Take and unlimited access to all of Bloomberg.com, subscribe today at Bloomberg.com slash podcast offer. If you like the episode, make sure to subscribe and review The Big Take Asia wherever you listen to podcasts. It helps people find the show. Thanks for listening. See you next time.

21:44This coffee shop running smooth. Thanks to genius from global payments, instant transactions, effortless inventory and synchronized operations, big league reliability for any business. That's genius. This is Robert Smith from business history. If you're listening to this, there's a good chance you're a small business owner. And like every small business owner, you started with a dream to do what you love and watch it grow. What you probably didn't dream about? Keeping up with cyber threats. That's where MasterCard can help with access to tools that help identify cyber threats to better protect your business.

22:19Building your dream business? Priceless. For cybersecurity in a changing world, there's MasterCard. Learn more at MasterCard.com slash small business. It's time to plan ahead and make sure your brand is showing up in ways that can have an impact. 4imprint's promotional products are designed to work as hard as you do and make a lasting impression. From quality apparel, including exclusive brands, to drinkware, tech, and totes, they've got thousands of options to fit your brand and budget. Plus, you get free samples, expert help, and their 360-degree guarantee. So you can be 4imprint certain everything shows up just right, right on time.

22:58Explore more at 4imprint.com. 4imprint. 4certain.