Indian workers record their daily activities to train artificial intelligence systems and robots, earning extra income while potentially helping develop technology that could threaten their livelihoods. From stitching garments and cutting hair to doing household chores, workers capture their movements on camera so technology companies can teach robots to perform physical tasks.
The growing industry presents India with an opportunity to participate in AI development. However, it also raises concerns about job security, workers’ bargaining power, and whether the country will benefit from the technology it helps build.
BBC’s report mentioned Mohammed Ashim, a 22-year-old barber in Bengaluru, who wears a head-mounted camera for five or six hours daily to record his hand movements while cutting hair. Rashdan Technologies collects the footage and supplies it to technology companies and research organizations.
The company pays USD 5 to USD 10 for eight to 10 hours of recorded work. Textile worker Mohammed Arman and science teacher Jolly Arifur also record their activities to earn extra income.
The recordings help train physical AI systems, enabling robots to learn real-world tasks by observing human movements, coordination, and force application.
Also Read: UAE-Based CNTXT AI Expands to the US With USD 300 Million in Revenue
As robots can perform increasingly complex physical tasks, workers in labor-intensive sectors could face greater competition from machines. Manufacturing, textiles and other industries that depend on human coordination and repetitive movements may be particularly exposed as automation advances.
The concern is that workers could help develop systems that eventually reduce demand for their skills. For people dependent on daily wages, the prospect of losing employment could outweigh the short-term financial benefits of recording their work.
Bino Paul, a senior professor at the Tata Institute of Social Sciences in Mumbai, warned, “By the time workers realize what they have participated in, things may have gone too far.”
India also risks remaining at the lower end of the global AI supply chain. Local companies collect and process footage, but technology firms developing advanced AI systems could capture much of the value it generates.
Manish Agarwal, founder of Humyn Labs, said, “India can't afford to stay a footage supplier.”
The country needs to build capabilities to process, validate,, and structure training data rather than relying primarily on recording human activities. Stronger safeguards and government regulation will also be important as the industry expands.
Without these steps, India could help accelerate automation while struggling to secure the economic benefits or protect workers whose skills make the technology possible.