Airbnb’s AI Coding Push

Airbnb’s AI adoption has reached a new threshold, Benzinga reported Thursday, with AI agents now responsible for 60% of all new code written at the company during the first quarter of 2026.

Chesky’s Productivity Claim Raises Eyebrows

CEO Brian Chesky made the disclosure during the company’s Q1 earnings call. He told investors that a single engineer can now accomplish what once required roughly 20 people. AI agents operate throughout the development pipeline but remain under human supervision, Chesky said. He framed the shift as a resource multiplier, giving the team capacity to build tools for API partners that previously fell outside its bandwidth.

The earnings backdrop was strong. Airbnb posted quarterly revenue of $2.7 Billion, an 18% increase year-over-year. The company carries a market capitalization of approximately $84 Billion. Shares closed Thursday at $141.49, up roughly 0.73% on the session and about 6% higher year to date.

AI Customer Support Also Gaining Ground

The productivity gains extend beyond engineering. Airbnb’s AI-powered customer support tool now resolves 40% of inquiries without escalating to a human agent. That figure was 33% earlier in the year, representing a meaningful jump within a single quarter. The trend suggests AI is compressing headcount requirements across multiple business functions simultaneously.

A Broader Race Across the Industry

Airbnb’s announcement arrives amid a wider sprint among technology firms to automate software development. Dario Amodei, CEO of Anthropic, said at the World Economic Forum in January that some of his own engineers had already stopped writing code by hand. He predicted AI could take over the bulk of software engineering work within six to twelve months. Microsoft, Google, and Meta have each disclosed rising shares of AI-assisted code in recent earnings cycles.

Also Read: Microsoft Reports Record Quarter Fueled by Azure AI Demand

Chesky Still Sees Gaps in AI for Travel

Despite his enthusiasm for internal AI tooling, Chesky offered a more cautious view of AI products aimed at consumers. He argued no company has cracked AI-driven travel booking yet. Current chatbot interfaces carry too much text, lack intuitive comparison features, and are built for solo users. Most travel decisions, he noted, involve multiple people, making a single-player chat format a poor fit for the category.

The comments suggest Airbnb views AI as a back-end competitive weapon for now, even as the consumer-facing AI travel race accelerates elsewhere in the industry.

Also Read: Anthropic’s Dario Amodei on AI and the Future of Work

Read Next: How Big Tech’s AI Spending Is Reshaping the Cloud Market