What happened
DoorDash launched "Tasks," a new app and integrated features within its existing courier app, paying its 8-million-strong U.S. contractor workforce to generate AI and robotics training data. Couriers in select U.S. markets, excluding regulated areas like California, can now film household chores, record conversations in Spanish, photograph restaurant menus, or scan supermarket shelves. This data evaluates DoorDash's in-house AI models and those of partners across retail, insurance, hospitality, and technology sectors, expanding on existing initiatives like closing Waymo robotaxi doors.
Why it matters
Gig economy platforms are becoming critical AI data collection infrastructure, shifting model development unit economics. Procurement teams and founders gain access to real-world, physical data at scale, reducing reliance on synthetic data or dedicated data labelling firms. This follows Uber's "Digital Tasks" programme, indicating platform engineers must integrate distributed human-in-the-loop processes for continuous model improvement. This model introduces new operational complexities and ethical considerations regarding contractor classification and data privacy.
Subscribe for Weekly Updates
Stay ahead with our weekly AI and tech briefings, delivered every Tuesday.
