There Will Not Be One Interface Says Brian Chesky 

Airbnb has started deploying its most significant artificial intelligence initiative to date: an AI-powered customer support agent now handling half of all service interactions in the U.S., according to CEO Brian Chesky. The tool, introduced quietly in April, has already contributed to a 15% decline in users needing human support and is expected to roll out to the entire U.S. user base in the coming weeks.

Chesky shared the update during Airbnb’s first-quarter earnings call on Thursday, calling the AI assistant the “best customer service agent in all of travel.” The bot, trained on hundreds of millions of past support interactions, is the first step in what Chesky describes as a longer-term plan to integrate AI more deeply across Airbnb’s user experience.

“It’s going to get significantly more personalized and agentic over the years to come,” Chesky told analysts, signaling that the company’s AI roadmap is designed for gradual scaling rather than aggressive front-end deployment.

The company has so far resisted the race to launch generative AI-driven planning tools, a contrast to rivals like Expedia and Booking.com, which have promoted features such as automated itinerary creation and real-time travel updates. Airbnb, by comparison, is limiting its initial use cases to support automation and backend workflow improvements, citing concerns over accuracy and readiness in more complex applications like trip planning.

Chesky has previously likened today’s AI tools to the early internet era, useful, but not yet stable enough to power core user journeys. That stance has shaped Airbnb’s rollout strategy, which prioritizes foundational infrastructure and operational support.

The new customer service bot replaces Airbnb’s prior approach of redirecting users to help articles, instead offering direct resolution for common issues through a chat interface. It is currently limited to English-speaking U.S. users but will expand to additional markets and languages later this year.

According to Judson Coplan, Airbnb’s VP of Product Marketing, the assistant is being positioned as more than just a support channel. Over time, it will evolve into a personalized concierge, assisting with bookings, providing travel suggestions, and facilitating faster resolution through smart prompts such as “Cancel my booking” or “Change dates.”

The goal, Coplan said in a recent product briefing, is to create an AI layer that can eventually handle both reactive support and proactive planning.

While Airbnb has not disclosed technical details, the assistant appears to be built on proprietary systems rather than relying entirely on external LLM providers like OpenAI or Anthropic. Chesky emphasized that the product is trained on a dataset unique to Airbnb comprising billions of data points from user interactions over the years.

The AI initiative coincides with a broader product overhaul. In May, Airbnb introduced a redesigned app with three core sections: Stays, Services, and Experiences. This update represents the company’s most expansive push beyond lodging since its founding.

Services now available on the platform include in-home chef-prepared meals, massages, personal training, photography, and grooming offered in 100 cities across eight countries. Users can book these services during their stay or independently, marking a shift toward Airbnb becoming a transactional platform for lifestyle offerings, not just accommodations.

The Experiences category launched in a limited form several years ago has been expanded to 19 categories across 1,000 cities globally. These include cooking classes, cultural tours, outdoor activities, and wellness events. Airbnb has also introduced a new sub-brand, Airbnb Originals, featuring high-profile collaborations with chefs, artists, and athletes to create exclusive sessions.

For example, users can book a pastry workshop at French Bastards bakery in Paris or a beach volleyball session with Brazilian Olympian Carolina Solberg.

To ensure quality, Airbnb is vetting service providers based on their experience, licenses, and digital presence. The company takes a 15% commission from services and 20% from experiences, although users see a single final price at checkout.

“These experiences and services are a way to experience a city like a local,” Coplan said. “They’re natural extensions of home stays.”

Chesky’s commitment to long-term reinvention appears rooted in a pivotal moment during late 2023. That November, as OpenAI CEO Sam Altman Chesky’s longtime friend and fellow Y Combinator alumnus was abruptly removed and then reinstated, Chesky said he found himself reflecting on Airbnb’s trajectory and future.

He spent the Thanksgiving weekend writing a new vision document for the company, which he presented to employees after the break. In that internal memo, Chesky outlined a plan to reposition Airbnb as a global platform not just for lodging, but for “living anywhere”—with travel, lifestyle services, and localized experiences as core pillars.

Despite Airbnb’s strong brand association with short-term rentals, Chesky acknowledged that “Airbnb is a noun and a verb” he said the company must evolve beyond its original product to remain competitive and relevant in a shifting market.

“I’d rather fail trying something new than succeed in doing the same thing over and over again,” he told an interviewer.

The company has also begun testing social features within the app, including group chat functionality for users who share trips or activities. Early data from past profile enhancements has shown a 15x increase in completed user profiles, which the company believes can improve engagement and personalization.

While Chesky supports the integration of AI, he pushed back against simplified visions of the future in which a single AI interface or device dominates consumer behavior.

“There isn’t going to be one chatbot, one app, or one device,” Chesky said. “Every application wants a different interface.”

He referenced the iPhone launch as a guiding design philosophy, noting that static keyboards were replaced by dynamic interfaces suited to specific apps. Airbnb, he suggested, is building toward a similar multi-interface environment, where AI augments interactions across several touchpoints rather than centralizing them.

Industry watchers say Airbnb’s strategy offers a contrast to the “AI maximalist” position adopted by some competitors, particularly those attempting to collapse the user journey into a single generative interface. While others experiment with all-in-one AI travel agents, Airbnb appears focused on building modular, context-specific AI tools grounded in operational scale.

The company did not break out AI investment figures in its earnings, but its Q1 results showed revenue growth alongside a cautious forecast on booking volume due to broader economic uncertainty. Shares remained relatively flat following the call.

“The future’s not going to be simpler,” Chesky said. “It’s going to be more complex. And we’re going to need more tools, not just one.”