From Ride-Hailing to Labeling: Inside Uber AI Solutions’ Quiet Rise

Uber is making a determined push into the business of AI training data, a market suddenly shaken by Scale AI’s move to sell a 49% stake to Meta. The ride-hailing and delivery giant has been quietly building a platform — originally called “Uber Scaled Solutions” — that now operates as “Uber AI Solutions,” matching thousands of clickworkers around the world with projects ranging from image tagging to legal document review.

The platform, which went live in November 2024, rides on Uber’s core strength: a global network of flexible on-demand workers. General Manager Megha Yethadka told Forbes that Uber is leveraging its experience as “the platform of choice for flexible work” to supply high-quality annotated datasets. The company already counts autonomous vehicle developer Aurora and Niantic among its early customers, and says the number of clickworkers on the platform has doubled since the start of the year.

Uber is not just selling finished datasets. It plans to license its internal project-management tools and its contractor network to clients, and is rolling out automation that lets customers describe data needs in plain language — with the system handling task distribution, workflow setup and quality assurance. The ambition is clear: to build a managed service around AI data labelling that rivals purpose-built annotation firms, all while being able to point to Uber’s $175 billion market cap as a stability guarantee.

Why the Scale-Meta Deal Is Reshaping the Data Annotation Market — and Where Uber Fits

The Scale-Meta Deal Jolts the Market

The catalyst for Uber’s accelerated push is the unravelling of Scale AI’s market dominance. By selling a near-majority stake to Meta, Scale has created both strategic and commercial friction. Leading customers such as OpenAI are already distancing themselves, and Google plans to re-evaluate its relationship. The spectre of one client — Meta — now effectively owning a big piece of Scale makes many AI firms uneasy about sending sensitive training data to that platform. Uber’s pitch as a neutral, independent alternative lands directly in that vacuum.

Uber’s Gig-Economy Muscle Meets AI

Uber’s entry is not a lightweight startup experiment. The company is now in over 30 countries, up from an initial five markets including the US, Canada and India. Top clickworkers earn $20 to $200 an hour depending on task complexity, typically working three to four hours a day — a model that mirrors Uber’s existing driver flexibility. “We’re a product and operations company that knows these processes ourselves,” Yethadka said, underscoring the ability to scale on-demand labour pools. Financial heft matters too: with 2024 revenue of $43.9 billion, Uber does not rely on venture capital in a space crowded by smaller players like Mercor, Turing and Invisible Technologies.

Quality and Talent Will Define the Leaders

Yet rivals caution that success in data annotation is increasingly about specialist expertise, not just volume. “Data annotation is shifting toward highly qualified work,” said Brendan Foody, CEO of Mercor, pointing out that Uber needs to build a deep talent network of subject-matter experts. The complexity of tasks — spanning STEM fields, coding, and legal domains — raises the bar for vetting and managing a global workforce. Uber’s ability to maintain quality while scaling will ultimately determine whether its platform becomes a default choice for enterprises or remains a supplementary supplier.

How Uber, Upstarts, and Scale AI Navigate a New Data Services Battlefield

  • For Uber: The rebranding to “Uber AI Solutions” and expansion of licenceable tools need to be paired with demonstrable quality benchmarks — especially in specialised fields like law and STEM — to win trust from large AI labs wary of data leakage. The upcoming rollout of the plain-language project interface could be a differentiator, but only if accuracy matches the ease of use.
  • For Scale AI: The Meta deal creates an urgent need to retain customers who see a conflict of interest. Scale may have to offer stronger data governance and independence guarantees, or even spin off certain operations to reassure partners.
  • For upstarts like Mercor, Turing and Invisible Technologies: Uber’s financial muscle is a direct threat, so differentiation must come from deep vertical expertise and faster turnaround on complex, domain-specific annotations. Partnerships with niche data owners could provide a moat.
  • For enterprise buyers: The shake-up in the annotation market means new negotiating leverage. Testing Uber AI Solutions against incumbents on a small project basis in late 2026 could provide cost and quality benchmarks before committing to longer-term contracts.

Risk & Opportunity Assessment

Commercial RiskMediumUber AI Solutions faces revenue concentration if enterprise customers consolidate around one or two providers post-Scale-Meta deal; however, the current fragmentation of the market offers a window for growth.
Competitive RiskHighThe data annotation sector is crowded with nimble startups (Mercor, Turing) and the incumbent Scale AI still retains deep client relationships and technical infrastructure, while Meta's new 'Superintelligence Lab' could also influence where data labelling budgets flow.
Regulatory RiskLowData privacy and cross-border transfer rules apply, but no imminent regulatory action specific to the annotation market is signalled; Uber’s existing compliance infrastructure from ride-hailing may help mitigate baseline risks.
Reputation RiskMediumUber’s history of gig-economy controversies around worker classification and pay could resurface as clickworkers are scrutinised; any quality lapse in AI training data may also attract negative attention given the critical nature of the datasets.
Technology DisruptionMediumAdvances in automated labelling and synthetic data generation could reduce the need for human annotators, but for the near term complex, domain-specific tasks still require expert human input, giving Uber’s managed-service model a window.
Commercial OpportunityHighThe Scale-Meta deal has created a significant trust gap that a well-capitalised, neutral platform can exploit. Uber’s global gig workforce and plans to license its entire infrastructure open a pathway to become a default data annotation partner for AI firms, with tangible revenue diversification beyond mobility and delivery.