US-based food-delivery company DoorDash is now paying its drivers to train AI. The company has introduced a new feature called Tasks, where drivers will get incentives to complete small real-world activities like taking photos of store shelves, checking product availability, recording videos, or verifying menus at restaurants. The company will then use the data collected through these tasks to train artificial-intelligence systems and improve real-world data for businesses.
According to DoorDash, the new Tasks feature is part of its plan to grow beyond food delivery into logistics, automation, and real-world data services. With this feature, the company will pay drivers, known as Dashers, for completing short activities such as taking photos of store shelves, checking product availability, recording videos, or verifying menus at restaurants. The company says these tasks will help businesses understand what is happening on the ground in real time, something that is otherwise difficult to track at scale.
The new Tasks option will appear inside the Dasher app just like a delivery request. Drivers can choose to accept these jobs between deliveries or complete them separately. The pay for these drivers will depend on the effort required and the complexity of the activity, and the details are shown before accepting the task. The company has also tested a separate app where workers may be asked to record conversations, film everyday activities, or capture videos that can be used to train AI and robotics systems.
Data from Tasks to train AI
The company says some of these tasks are directly linked to improving artificial-intelligence models. “This data helps AI and robotic systems understand the physical world. Pay is shown upfront and determined based on the effort and complexity of the activity,” says DoorDash in its official blog post.
In certain cases, drivers can also submit video clips, record speech, or film household actions so that AI systems can learn how humans perform everyday tasks. This collected data can then be used by DoorDash and its partners in industries such as retail, insurance, hospitality, and technology to train AI models and improve automation tools.
DoorDash is tapping into its large network of drivers to meet the growing demand for real-world data needed to build smarter AI systems. And DoorDash is not alone. Earlier, Uber launched a “Digital Tasks” pilot programme in 12 Indian cities, including Delhi, Mumbai, and Bengaluru, as well as in the US.
Using this feature, Uber allows drivers and couriers to earn extra income during downtime by completing tasks such as labelling images, classifying text, counting objects, digitising receipts, recording audio samples, and uploading photos of specific items or restaurant menus. According to the company, these tasks feed data into Uber AI Solutions, which improves internal services and also provides training data to enterprise clients.
Amazon is also using its delivery drivers to train AI through technologies such as AI-powered smart glasses. These glasses can capture real-world data as drivers navigate neighbourhoods, scan packages, and identify hazards. This information can be used to train autonomous delivery systems and improve route-optimisation algorithms.
