Uber’s chief operating officer Andrew Macdonald says the company’s use of AI coding assistants like Calude Code and ChatGPT has increased the internal productivity by about 25%, but he also admits there is no direct link yet to delivering more useful consumer-facing features. According to a report by Fortune, speaking on the Rapid Response podcast, Macdonald explained that while AI adoption has sped up development, it’s difficult to measure how those gains translate into new features for riders and drivers. “That link is not there yet,” he said. “Maybe implicitly there’s more that is getting shipped, but it’s very hard to draw a line between one of those stats and ‘Okay now we’re actually producing like 25% more useful consumer features.’”
Cost concerns
Uber reportedly exhausted its 2026 AI coding tools budget just four months, after incentivising employees to use AI through an internal leaderboard. The surge in usage highlights a broader challenge in enterprise AI adoption: costs are rising even as per-unit AI pricing falls. Macdonald cautioned, “If you’re not actually able to draw a direct line to how many useful features you’re shipping to your users, that trade becomes harder to justify.”Other tech firms are facing similar dilemmas. Microsoft has scaled back Claude Code licenses, shifting engineers to GitHub Copilot CLI, while Duolingo’s CEO Luis von Ahn has reversed earlier optimism about AI’s role in replacing human tasks.Uber CEO Dara Khosrowshahi recently said about 10% of the company’s committed code is now built by autonomous agents, and AI tools are being adopted across legal, marketing, and engineering teams. “We think it’s creating employees with superpowers,” he said.
Uber CEO Dara Khosrowshahi admits that company is slowing hiring and the reason is AI
In related news, recently Uber CEO Dara Khosrowshahi During the company’s first-quarter 2026 earnings call, Khosrowshahi explained that the productivity gains offered by AI are significant enough that they justify a shift in spending from human hires to technology.“I think you should just look at AI as an accelerator. For us, for every company, it means that our investment in AI tools and infrastructure is increasing. That will be offset by slower headcount growth,” Khosrowshahi said.Uber has integrated AI into its operations, from legal and marketing teams to its core software engineering. The CEO noted that these tools are not just assisting staff but are creating “employees with superpowers”.He revealed that roughly 10% of Uber’s code changes are now produced by autonomous AI agents, pointing out that while AI writes the code, human employees still review and test it before it is officially added to the company’s repository. According to Khosrowshahi, this has increased the number of code commits per engineer, allowing the company to move faster in building and maintaining its systems.
