Gig platforms are increasingly treating their delivery workforces as a deployable sensor network, capable of generating the real-world data that AI systems require at scale.
DoorDash announced a stand-alone “Tasks” app that pays its delivery couriers — called Dashers — to complete assignments designed to improve AI and robotic systems. According to the announcement, workers can earn money by filming everyday activities or recording themselves speaking in another language, with pay shown upfront and calibrated to the effort and complexity involved. “This data helps AI and robotic systems understand the physical world,” the company wrote in a blog post.
The footage and audio collected will be used to evaluate both DoorDash‘s own AI models and those built by partners across the retail, insurance, hospitality, and technology sectors. One specific task, according to the report, asks a courier to film their hands washing at least five dishes while wearing a body camera, holding each clean dish in frame for a few seconds before moving to the next.
Beyond the Stand-Alone App
Alongside the new application, Dashers will also see Tasks listed directly inside the existing Dasher app. These include photographing restaurant menu items to help businesses showcase their offerings, or capturing hotel entrances so other drivers can locate drop-off points. DoorDash‘s existing partnership with Waymo — in which couriers are paid to close the doors of autonomous delivery vehicles — is also categorized as a task within the platform.
Ethan Beatty, general manager of DoorDash Tasks, framed the scale of the opportunity directly: “There are more than 8 million Dashers who can reach almost anywhere in the U.S. and who want to earn flexibly beyond delivery. That’s a powerful capability to digitize the physical world.”
Both the in-app Tasks and the stand-alone application are currently available in select U.S. markets, with California, New York City, Seattle, and Colorado excluded from the launch. The company says it plans to expand into additional task types and countries.
A Pattern Forming Across Gig Platforms
DoorDash is not alone in this approach. Late last year, Uber announced plans to let drivers earn supplemental income by completing small jobs such as uploading photos to help train AI models — a near-identical structure applied to a different fleet.
What both moves reflect is a logical extension of the gig model: workers who already move through the physical world, equipped with cameras and smartphones, represent a distributed data collection infrastructure that would otherwise be expensive and slow to build through conventional means. The monetization logic runs in both directions — platforms acquire proprietary training data while workers gain income streams that do not depend on active deliveries.
Photo by ANNIE HATUANH on Unsplash
This article is a curated summary based on third-party sources. Source: Read the original article
