Uber, the ride-hailing giant, is diversifying its platform beyond trips and deliveries by introducing a new revenue stream for US drivers: short "digital tasks" that can be completed in minutes, even during downtime like waiting for passengers or charging electric vehicles. Announced at the Only on Uber conference, this pilot program taps into the booming demand for human-assisted AI training data, positioning Uber as a versatile hub for flexible work.
The Digital Tasks Pilot: Earning on the Go
Starting later this fall, a dedicated "digital tasks" category will appear in the Uber Driver app for select participants. These micro-jobs, powered by Uber's AI Solutions Group, are designed for seamless integration into a driver's routine - requiring no additional apps or commitments. Each task typically takes 2-3 minutes and pays a fixed amount, deposited into the driver's balance within 24 hours.
Examples include:
- Data labeling for AI models: Uploading photos of everyday scenes, like cars or street signs, to help train computer vision systems;
- Uploading restaurant menus: Snapping and submitting images of local eatery menus, especially those in non-English languages;
- Recording voice samples: Speaking prompts in specific languages, accents, or dialects to improve speech recognition tools;
- Script narration: Reading short scenarios aloud for multilingual audio datasets.
Uber emphasizes that tasks won't appear while drivers are actively online and accepting rides, prioritizing road safety. The initiative builds on a successful beta in over 12 Indian cities, where drivers have already embraced similar opportunities. "We're making Uber the best platform for flexible work," said CEO Dara Khosrowshahi at the launch event, highlighting how these gigs address the "infinite workday" by monetizing idle moments - whether at home or in a parking lot.
Tapping into the AI Data Gold Rush
This move isn't just about driver perks; it's a strategic play in the exploding AI data labeling market. Companies rely on human annotators to "teach" models by tagging, categorizing, and validating vast datasets - essential for advancements in generative AI, autonomous vehicles, and more. Uber's vast network of over 6 million global drivers and couriers provides a ready-made, on-demand workforce, potentially disrupting incumbents.
The sector's scale is staggering. Scale AI, a key player in high-quality data annotation, reported significant revenue and projects even higher figures, with a recent investment valuing it at around $29 billion. Similarly, Surge AI, focused on reinforcement learning data, hit $1 billion in annual revenue and is eyeing a $1 billion raise at a $25 billion valuation. Uber's entry could capture a slice of this pie, serving clients like Aurora (self-driving tech), Niantic (AR gaming), and Luma AI (video generation) through its Uber AI Solutions division.
Bolstering Capabilities with Strategic Acquisition
To supercharge this venture, Uber acquired Belgian startup Segments.ai - a firm specializing in multi-sensor data labeling for robotics, drones, and autonomous systems. Segments.ai's expertise in LiDAR and sensor annotation tools will enhance Uber's offerings, reducing manual labeling efforts via cloud-based automation and smart suggestions.
Founders Otto Debals and Bert De Brabandere, along with the team, have joined Uber AI Solutions to scale these capabilities globally.
"This acquisition aligns with our decade-long work in data-driven AI for autonomy and safety," Uber stated, noting Segments.ai's client base and domain knowledge as key assets. The deal underscores Uber's shift from consumer rideshare to enterprise AI services, now available in 30 countries with tools for multimodal datasets.
Also read:
- What is PimpAndHost?
- Think Like a Whale: The Crypto Mindset for Unshakeable Success
- Hackers Hijack Official Dota 2 YouTube Channel to Pump Solana Meme Coin Scam
Broader Implications for Gig Workers and AI Ethics
For drivers, this means diversified income without the wear-and-tear of extra miles - potentially adding $0.50 to $1 per task, though Uber cautions these figures are illustrative. It's a response to feedback from the gig economy's front lines, where downtime often means lost earnings. Competitors like Amazon's Mechanical Turk offer similar crowdsourcing, but Uber's app integration could make it more accessible.
Yet, as AI training relies more on human labor, questions arise about fair pay, data privacy, and job displacement. Uber insists tasks are unrelated to its autonomous vehicle efforts, focusing instead on third-party clients. Still, with AI's rapid evolution, this pilot could redefine "side hustles" in a world where every uploaded photo or spoken word fuels the next tech breakthrough.
As Uber eyes expansion, the program signals a future where gig platforms evolve into multifaceted earning ecosystems—blending wheels, wheels-off work, and the invisible labor powering tomorrow's machines.
