Uber slows hiring as it invests in AI
On the companys first-quarter earnings call, CEO Dara Khosrowshahi said that roughly 10% of Ubers code changes are produced by autonomous AI agents, and that humans review that code before it is merged, Business Insider reports. Khosrowshahi described the tools as creating "employees with superpowers" while saying the company is "metering headcount growth" as it increases AI investment, according to Business Insider. Business Insider also reports that Ubers CTO said last month the company had already spent its entire 2026 budget for Claude Code, the Anthropic coding model.
What happened
On Ubers first-quarter earnings call, CEO Dara Khosrowshahi said that about 10% of the companys code changes are produced by autonomous AI agents and that human engineers still review changes before they are merged, Business Insider reports. Business Insider quotes Khosrowshahi saying the tools create "employees with superpowers" and that Uber is "metering headcount growth" while investing in AI. Business Insider also reports that Ubers CTO said last month the company had spent its entire 2026 budget for Claude Code.
Editorial analysis - technical context
Companies deploying AI-assisted development workflows typically use a mix of agentic tools, code-generation models, and human-in-the-loop review to reduce routine work while retaining safety controls. Observed deployments commonly focus on code scaffolding, refactor suggestions, and test generation rather than fully autonomous feature development. For practitioners, the 10% figure reported by Business Insider is a concrete peer datapoint for adoption level in a large engineering org; however, the continued human review noted on the call aligns with industry best practices for production safety and code quality.
Industry context
Wider reporting on large technology firms shows an accelerating trend of reallocating hiring growth or "metering" headcount while increasing investment in AI tooling. Industry observers have framed similar moves as cost-productivity tradeoffs where organizations buy productivity via tooling rather than linear headcount increases. Those patterns reshape engineering team workflows, onboarding, and tooling priorities across companies of similar scale.
What to watch
- •Adoption metrics: whether the share of AI-generated changes rises above 10% and how rollback/bug rates move.
- •Governance and safety: evidence of automated testing, static analysis, and human review coverage for agent-produced patches.
- •Budget signals: further disclosures about spending on models like Claude Code and whether vendor usage expands beyond coding assistants.
Scoring Rationale
This is a notable company-level signal: a large tech employer reporting **10%** of code from AI agents and slowing hiring for AI investment matters to practitioners evaluating tooling adoption and engineering productivity. It is not a paradigm-shifting model release, so the impact is moderate.
Practice with real Ride-Hailing data
90 SQL & Python problems · 15 industry datasets
250 free problems · No credit card
See all Ride-Hailing problems

