Uber Rolls Out ChatGPT Work to All 30,000 of Its Employees

Uber has taken what amounts to one of the broadest enterprise-wide AI adoption moves to date, announcing an expansion of its partnership with OpenAI that will make ChatGPT Work tools available to its entire global workforce — roughly 30,000 employees. The decision wasn't made all at once; it followed a limited pilot phase that demonstrated the tools' value extends well beyond engineering teams, a finding that appears to have settled the company's final direction.

From a 1,000-Employee Pilot to a Company-Wide Rollout

Uber began its journey with ChatGPT Work — the product OpenAI launched for enterprise use as an intelligent agent that can be integrated into daily workflows — through a pilot involving around 1,000 employees earlier this year. The company spread the pilot across multiple departments, including finance, operations, and product, rather than confining it to a single technical team.

Notably, the results showed that the overwhelming majority of early users — an estimated 80% — were not engineers. That particular data point deserves attention: the greatest value was realized by people without a programming background, reflecting how generative AI tools have lowered the barrier to entry for tasks that previously required a technical intermediary.

Practical, Not Experimental, Use Cases

According to Sumanth Sukumar, Uber's VP of Engineering, employees used the tool across a diverse range of tasks, most prominently:

• Analyzing complex operational data to extract trends and build actionable recommendations.
• Adapting training materials and translating them to suit the various markets and languages in which the company operates.
• Building custom interactive dashboards that each team uses to manage its day-to-day work.

These use cases share a common trait: all are moderately complex tasks that previously consumed substantial time or sat waiting in the queue of a data or engineering team's backlog. Shifting that capability into the hands of the employee is the core organizational gain being anticipated.

Testing "AI Pods" Within Teams

Alongside the broad rollout, Uber tested a different model it refers to internally as "AI pods," embedding roughly 30 engineers specializing in artificial intelligence within various teams across the company and giving them two weeks to automate core portions of those teams' workloads.

The lesson drawn from this experiment, according to Sukumar, is that the real opportunities emerge when technical expertise is combined with a deep understanding of the nature of the job itself. It's an observation worth considering for any organization planning a similar path: tools alone don't produce transformation — a precise understanding of what can genuinely be automated does.

Deploy First, Define Use Later

What stands out about Uber's approach is that it didn't predetermine how employees should use the tool. The stated philosophy is to make powerful capabilities broadly available, give teams room to experiment, and then observe the genuine work patterns that emerge.

The company said it will track how the technology is used, the kinds of workflows employees devise, and the extent to which these contribute to greater effectiveness — then evaluate those results to guide future deployment decisions. Uber did not disclose the spending associated with the expansion, though it noted that OpenAI's teams worked closely with it during the integration phase.

The Broader Context: The Battle for the Enterprise Market

This announcement comes against a clear competitive backdrop. OpenAI has moved quickly to expand its enterprise business in the face of prominent rivals in the sector, after years of being known primarily as a consumer-facing brand. The company has succeeded in building an enterprise customer base at a faster pace than many of its competitors.

Having a client of Uber's scale — with its wide geographic footprint and departmental diversity — gives OpenAI more than just a commercial contract; it provides a reference model it can showcase to other organizations hesitant about moving from limited pilots to full-scale adoption.

Key Takeaways

Uber's experience offers three practical signals for any organization weighing a similar move: start with a cross-departmental pilot rather than a narrow technical one; expect that the user who benefits most may not be an engineer; and base decisions on actual usage patterns rather than imposing predetermined scenarios. At this stage, what matters most isn't having the tool — it's the ability to measure its impact accurately and translate that into tangible workflow improvements.

✦ بقلم فريق دروب أيديا