Thousands of factory workers in Pakistan are becoming part of the global race to build smarter robots by doing something unusual: performing their normal jobs while cameras and sensors record exactly how they move.
Pakistani startup World Context is collecting this real-world movement data and turning it into datasets that robotics companies and researchers can use to teach machines how to perform physical tasks. The company says it currently collects data from around 8,000 workers.
Founder and CEO Sherdil Ali Khan says participating workers are paid roughly twice their previous wages.
World Context works with people across industries including textiles, automotive manufacturing, electronics, tooling, logistics, health care, heavy industry and specialist manufacturing.
Workers continue doing tasks they already know, such as gripping tools, fitting components together, pressing controls and adjusting objects.
They wear equipment including smart gloves and cameras mounted on their heads, wrists and feet. These devices record the small movements involved in completing each task.
Modern AI can learn from huge amounts of text, images, and other digital information, but robots face a harder problem: they need to understand the physical world.
A robot has to learn things humans often do without thinking, such as how tightly to grip something, how much force to apply, how a surface feels, and how to react when an object does not move exactly as expected.
Factory workers already possess these skills.
World Context records those actions so robotics developers can use them as examples when training machines.
The company says every recording is covered by a written agreement, checked for quality, and documented to show where the data came from. It can also produce custom datasets for companies that need examples of specific tasks or working environments.
Khan says World Context is trying to create a data supply chain connecting Pakistan with robotics companies in the United States and elsewhere.
The company says US robotics companies have already used Pakistani factory data.
Pakistan’s large industrial workforce could give it an advantage, as workers have experience across many manufacturing environments that are difficult to replicate through computer simulations.
The model creates an unusual relationship between workers and automation.
Robots may eventually automate some of the same physical jobs these workers currently perform. Yet before machines can do those jobs effectively, companies need humans to demonstrate how the work is done.
That means Pakistani workers could first benefit financially from the same technology that may eventually reduce the amount of human labor required for some industrial tasks.
World Context is effectively turning human skill itself into AI training data, giving Pakistan another possible route into the rapidly expanding global robotics industry.
Get the latest tech news, telecom insights, and product launches wherever you prefer.
Add ProPakistani to Preferred Sources and see more of our stories in Google Search and Top Stories.
Technology and Automotive Specialist covering the latest cars, smartphones, AI breakthroughs, and...
Shares
