Uber’s App Becomes an AI Training Ground

Uber is transforming its popular ride-sharing and delivery application into a multifaceted platform for flexible income, now enabling its U.S.-based drivers and couriers to earn additional money by completing brief digital assignments designed to advance artificial intelligence technology. These new microtasks involve activities like recording voice clips, taking and submitting photographs, or uploading documents in specific languages. For instance, a driver might receive a prompt asking them to upload images of vehicles or record spoken phrases in their native dialect. Another task could involve photographing a Spanish-language menu, potentially earning the worker up to one dollar per completed assignment.

This strategic initiative places Uber’s vast network of freelance workers in direct competition with established data-labeling services such as Scale AI and Amazon’s Mechanical Turk. These platforms are essential for the AI industry, employing human workers to annotate and categorize the massive datasets that train sophisticated machine learning models. Traditionally, this labor-intensive work is outsourced to lower-cost international markets, but Uber’s pilot program brings these opportunities directly to its American workforce.

The announcement is a key part of Uber’s broader corporate mission to become the premier destination for flexible work, a vision CEO Dara Khosrowshahi recently emphasized at an event in Washington, D.C. Alongside the AI tasks, the company is implementing a suite of app enhancements intended to improve the working experience for its drivers and couriers, making it more intuitive, efficient, and equitable.

Uber is no stranger to leveraging its workforce for AI development. The company has previously utilized what it terms “human-in-the-loop” systems, which integrate human judgment with automated processes. A recent acquisition of the Belgian startup Segments.ai further signals its commitment to expanding its data-labeling capabilities. After testing similar microtasks with drivers in India, the company is now launching this pilot program in the United States.

Whether drivers will enthusiastically embrace these new tasks is an open question. Many already voice concerns about inadequate earnings, citing the company’s substantial commission on rides and deliveries. The ongoing debate over driver classification adds another layer of complexity. Uber maintains that its drivers are independent contractors, a status that excludes them from standard employee benefits like minimum wage guarantees and health insurance. Conversely, many drivers argue that the level of control exerted by Uber’s algorithm effectively makes them employees.

Beyond AI tasks, Uber is revamping the trip offer cards that drivers see when a new ride request comes in. The updated interface provides more time and detailed information, allowing drivers to make better-informed decisions before accepting a job. For delivery couriers, a simplified on-trip experience is being introduced to clarify multi-order pickups and drop-offs, complete with alerts for items that are frequently forgotten.

A newly designed heatmap offers drivers greater transparency into areas with high passenger demand. The color-coded system uses red to indicate the shortest anticipated wait times, followed by orange and yellow. Purple zones highlight areas where surge pricing is active, and the map displays historical data on average wait times. Drivers commuting from home to a busy urban center can now select a route that either gets them there fastest or one that maximizes their potential to pick up fares along the way.

Safety features are also receiving significant upgrades. Following its initial U.S. launch last July, the Women Rider Preference feature is expanding to additional cities like Baltimore, Minneapolis, and Seattle. This tool allows female drivers to set their app to only receive trip requests from women passengers, and vice versa. Uber reports that this feature has already been used on more than 100 million trips, with a quarter of women drivers activating it weekly and over half using it for the vast majority of their rides.

Drivers now have the ability to set minimum rider ratings based on their personal comfort levels, enabling them to toggle this filter on for late-night shifts and off during the day. Combined with enhanced rider verification processes, these tools are designed to provide drivers with greater peace of mind.

Addressing long-standing grievances about account deactivations, Uber is introducing new measures focused on fairness. The company acknowledges that losing platform access creates significant hardship and is now working to limit restrictions to specific earning opportunities rather than imposing a full ban for minor infractions. For example, a reported issue with an alcohol delivery might restrict a driver from future grocery deliveries but still allow them to accept food delivery or ride-sharing trips. Serious safety violations, however, will still result in complete deactivation.

In disputes, drivers will now have the opportunity to present their side of the story before a final decision is made. Riders who submit demonstrably false reports about a driver will themselves face the risk of being removed from the platform.

Finally, a new Delayed Ride Guarantee ensures that drivers receive additional compensation when a trip is prolonged due to customer delays or unforeseen circumstances like heavy traffic. Uber is also strengthening its tipping reminders for passengers, expanding these prompts to iPhone Live Activities to encourage more riders to show their appreciation financially.

(Source: The Verge)