Uber’s AI spend has outpaced its budget, burning through the entire 2026 allocation for coding tools in just four months. The company incentivized engineers with a leaderboard that tracked total Claude Code usage, a move that quickly exhausted the forecasted funds. Executives now question whether the surge in AI‑driven features actually benefits riders, across its global platform, from surge pricing to route prediction.

President and COO Andrew Macdonald told the Rapid Response podcast that linking the rise in Claude Code usage to concrete consumer improvements remains elusive. He noted the difficulty of tying internal AI metrics to a measurable increase in useful features, a gap that makes the spend harder to justify. Uber’s R&D outlay rose 9% YoY, a significant increase, reaching $951 million in Q1 2026.

The budget blowout mirrors a broader enterprise AI dilemma as firms grapple with rising usage costs despite falling per‑token prices. Competitors like Microsoft are scaling back Claude Code licenses in favor of GitHub Copilot CLI, while Anthropic shifts to usage‑based pricing. Uber continues to pour resources into autonomous‑driving projects, betting that long‑term gains will outweigh near‑term overruns in the near term.