What did DoorDash Tasks app do?

DoorDash launches “Tasks” for AI training

DoorDash has introduced a new standalone “Tasks” app option that pays delivery couriers in some markets to submit short video clips and complete other assignments intended to train AI models.

How it works in practice

The reporting describes a real-world loop: couriers can be asked to record video clips as they complete tasks during normal work, and then DoorDash pays them for participation. The “content” is framed as training material for AI systems—suggesting use cases like improving video understanding, labeling, or teaching models how certain actions look on camera.

Why it’s part of a broader AI labor shift

This move fits a larger trend in the AI economy: companies need large amounts of labeled data, and some of the work is being distributed to gig workers. Rather than traditional annotation services, platforms can recruit people already performing repetitive, camera-friendly activities.

What’s unclear

The summaries do not specify:

• which specific AI model families the data is for,
• what consent and retention policies are for the footage,
• or how training labels are verified.

For users and workers, the key practical concern is whether the tasks add friction to delivery work, and whether compensation and data handling terms are clear.

Either way, it’s a concrete example of how generative AI development is increasingly tied to online labor platforms, turning everyday activities into training inputs.