Uber’s gig within gig allows drivers to make money labeling AI data

Uber has launched a pilot program in the US that allows its drivers to complete data labeling tasks during downtime, expanding on a similar initiative already running in India. The move positions Uber to become a major player in the AI training data market by leveraging its global workforce of drivers and delivery workers to help enterprises scale their artificial intelligence efforts.

What you should know: Uber’s AI Solutions division is testing a program where select US drivers and couriers complete digital tasks like recording speech samples, submitting multilingual documents, and uploading images.

• The pilot builds on a successful program launched last month across 12 Indian cities, where drivers use downtime to handle tasks including image classification, text analysis, audio transcription, and receipt digitization.
• Megha Yethadka, global head of Uber AI Solutions (formerly Scaled Solutions), announced the US expansion at Uber’s Only on Uber event.

The big picture: Uber is transforming from a ride-sharing company into a comprehensive data and logistics platform that can compete directly with specialized AI training providers.

• The company’s June announcement described its AI platform as offering “customized solutions for building smarter AI models and agents, global digital task networks, and tools to help companies build and test AI models more efficiently.”
• Uber AI Solutions provides “realistic task flows, high-quality annotations, simulations, and multilingual support, helping AI agents understand and navigate real-world business processes.”

Why this matters: The initiative introduces significant competition to the human-in-the-loop data labeling market, which has been dominated by specialty vendors and remains costly for many organizations.

• Shashi Bellamkonda from Info-Tech Research Group, a technology advisory firm, called it a “smart and strategic move” that could bring “more flexibility and pricing pressure to a market that is still too costly and resource-intensive for many organizations.”
• The approach mirrors how CAPTCHA systems train AI models, allowing Uber to generate valuable training data internally while potentially offering annotated datasets to other companies.

What experts think: Industry analysts see this as evidence that data preparation is becoming more commoditized and strategic for enterprise AI initiatives.

• “Uber is a data and logistics company first, and this move uses those strengths in a way that could generate valuable and diverse training data,” Bellamkonda observed.
• He advised CIOs to explore whether their own operational data could be used for proprietary data-labeling initiatives and to assess Uber’s future offerings against existing providers.
• “CIOs should begin viewing these functions as strategic procurement categories rather than just niche technical tasks,” Bellamkonda noted.

Strategic implications: The expansion demonstrates how companies can leverage existing workforce infrastructure to create new revenue streams while addressing enterprise AI challenges.

• Uber’s global reach provides access to diverse, multilingual datasets that could be particularly valuable for companies seeking internationally representative training data.
• The model offers potential cost savings for enterprises struggling to align AI investments with clear business value.