With a smartphone strapped to her head, 25-year-old Indian housewife Nagireddy Sriramyachandra spends part of her day filming routine household activities such as slicing mangoes, washing utensils, and carrying out kitchen chores.
According to AFP, these seemingly ordinary recordings are helping train artificial intelligence-powered robots that could eventually perform similar household tasks autonomously.
Sriramyachandra, who lives in Chennai in Tamil Nadu, earns a little over two dollars for every hour of footage she records. While the work may appear simple, the data she generates is considered highly valuable by global technology companies seeking to teach robots how humans move and interact with real-world environments.
"Who else will give you 250 rupees an hour just for doing housework?" said Sriramyachandra from her kitchen in Chennai in southern India's Tamil Nadu state.
"I may get a robot myself in the future," she added.
According to AFP, Sriramyachandra is among a rapidly expanding workforce of thousands of AI system trainers across India. Their work forms a crucial part of efforts to develop advanced robotic systems capable of navigating complex physical environments.
Unlike AI chatbots and image generators that rely on vast amounts of digital information, robots require real-world training data to understand human actions and surroundings.
AFP reported that developers increasingly rely on first-person recordings, known as "egocentric data," to help machines learn how to replicate human behaviour and movement.
To generate such data, AI trainers use a range of equipment including smart glasses, head-mounted cameras, smartphones, and motion sensors. Some contributors work from their homes, while others operate from factories and specialised recording studios designed specifically for AI training projects.
Photo Credit: AFP
"It blares 'hands not detected' when I'm not recording properly," said Sriramyachandra, who sends recordings via a special app to the AI data company Objectways.
The company, which operates offices in both India and the United States, serves several Fortune 500 clients and collaborates with Amazon SageMaker, a platform used for developing machine-learning models.
Booming Humanoid Robot Industry
According to AFP, the global humanoid robot market is witnessing rapid growth. Investment bank Morgan Stanley estimates that more than one billion humanoid robots could be in use worldwide by 2050, with the majority deployed in industrial and commercial sectors.
"Folding clothes, coffee making... cooking a very specific thing, sandwich making," Objectways head Ravi Shankar said, listing videos requested by clients.
"Some jobs are supposed to be taken over, so humans can go and do better things."
In India, the emerging field of spatial AI is creating employment opportunities, at least in the short term. Shankar, who is based in the United States but grew up in Tamil Nadu, recruits extensively from the state, which is widely recognised as one of India's major technology hubs.
During a visit to a textile factory in Karur, Tamil Nadu, AFP observed workers attaching labels to caps and ironing cloth bags while simultaneously wearing smart glasses and head-mounted cameras supplied by Objectways. Their recordings are used to train AI systems to understand industrial and manufacturing tasks.
Photo Credit: AFP
India has increasingly positioned itself as a global hub for AI data collection, annotation, and processing services. Experts believe demand for such work is likely to grow significantly as AI applications become more sophisticated.
"It's likely that these data collection services will increase", said digital labour expert Aditi Surie, from the Indian Institute for Human Settlements in Bengaluru.
Balancing Opportunity and Automation Risks
As India accelerates its AI ambitions, policymakers are also examining the impact of automation on employment. According to AFP, government think-tank NITI Aayog has highlighted concerns that discussions surrounding artificial intelligence often focus primarily on white-collar workers while overlooking the country's vast informal workforce.
Government think-tank NITI Aayog said that most discussions around artificial intelligence and labour "focus on white-collar professionals and predict an almost certain loss of jobs in the segment" without urgent action.
"Little attention, if any, is paid to how AI can serve India's 490 million informal workers, the very people who form the backbone of our economy," it said in a report released ahead of a global AI summit in India this year.
The think-tank has studied how AI could impact a wide range of occupations, including cobblers, sewer cleaners, farmers, flower sellers, and tea vendors. Policymakers are attempting to understand both the opportunities and challenges that automation may create for these professions.
Among those contributing data to AI systems is 55-year-old Ponni, who has spent more than a decade making flower garlands on a roadside in Bengaluru, often referred to as India's Silicon Valley.
According to AFP, she has also participated in AI training projects by recording her work while wearing a camera mounted on her forehead.
"The next generation... who might have to do work similar to mine - they will face a problem," Ponni said.
Life Behind the Camera
At one of Objectways' specialised studios, AI trainers spend hours filming themselves carrying out routine household activities in realistic apartment-like environments. According to AFP, the rooms are fully furnished and designed to simulate real homes, allowing AI systems to learn from diverse situations and settings.
After thousands of hours of recording, even elements such as wallpaper are changed periodically to provide clients with a wider range of visual environments.
"Today I sit here, tomorrow I stand there," said engineering graduate Rani N., 21, on a break from filming herself, once again, folding a towel.
Each recording lasts approximately four minutes, and she completes nearly 90 videos every day, capturing the same task from numerous positions and angles. Although she describes the work as manageable, she admits that the constant recording can feel overwhelming.
She says the job is "tolerable", but feels like she's always wearing a camera.
Elsewhere in the studio, workers arrange everyday objects such as pencil sharpeners, water bottles, and crayons into different patterns while recording them using depth-sensing cameras. These datasets help AI systems understand spatial relationships and object recognition.
Qanat Consulting Services, based in Andhra Pradesh and operating as an Objectways subcontractor, supplies recordings to around a dozen larger AI data firms. Chief executive Thaslim Pattan said some contributors wear motion-sensing devices on multiple parts of their bodies while carrying out assigned activities.
Some of its 2,000 contributors perform tasks with motion-sensor bands on their "wrists, hands and legs", chief executive Thaslim Pattan said.
Meanwhile, Bengaluru-based Humyn Labs focuses on collecting conversational data alongside video recordings. Contributors discuss assigned topics ranging from politics and current affairs to entertainment and lifestyle, helping AI systems learn speech patterns and natural language interactions.
According to AFP, despite growing concerns about automation, some industry leaders believe AI and humans will ultimately work together rather than compete against one another.
Manish Agarwal of Bengaluru-based Humyn Labs, not related to Objectways, records conversations as well as videos.
Agarwal denies that robots will steal jobs, believing that networks of humans and robots "will work together" one day, he said.
"A welder in India could be managing a welder-robot in Prague," he said.
As artificial intelligence continues to reshape industries worldwide, India’s growing army of AI trainers remains at the centre of a technological transformation—teaching machines to replicate human actions today while helping define the future relationship between people and robots tomorrow.