Brian Chesky Says AI Agents Need Their Own Operating System

Airbnb’s AI Search Update and the Limits of Chatbots

Airbnb introduced a new AI‑powered search feature this week as part of its fall update, marking a notable shift for a company that has moved cautiously on artificial intelligence. Co‑founder and CEO Brian Chesky has long argued that a pure chatbot interface is ill‑suited for travel discovery. Instead, he envisions a middle ground between today’s conversational bots and the static listings users currently see.

“I think that I’ve believed for a long time that a chatbot isn’t the right interface for e‑commerce. I think what you’re seeing today is also not the endgame for e‑Commerce or for travel or shopping. I think that there’s going to be some interface that is, you know, I don’t want to call it a midpoint, but something between a chatbot and what you see in the first version we shipped,” Chesky said.

He added that chatbots fail at browsing because they present only a few options at a time, forcing users to engage in multiple turns before reaching a result. Moreover, travel research shows people derive more pleasure from planning and anticipating trips than from the actual travel, so Airbnb wants to preserve the inspirational, library‑like experience of dreaming about destinations.

From Single‑User Bots to Multiplayer AI

Chesky also pointed out that chatbots are built for one person, lacking the collaborative nature needed when families or friends plan trips together. Over the next three to six months, Airbnb will experiment with interfaces that support “multiplayer” AI, allowing several users to interact with the same AI simultaneously.

While Airbnb rethinks its interface, other companies are exploring generative screens that an AI constructs on the fly. Startups such as Monogram and, more recently, Wabi are testing these dynamic interfaces. Chesky expects users will encounter a blend of deterministic (pre‑designed) and generative UI elements in the future.

Building an Agent‑Friendly Platform

Airbnb is also preparing for consumer AI agents—tools like Instinct and Muse that act on a user’s behalf. Chesky said these agents will require robust software infrastructure and that Airbnb aims to become more agent‑friendly, potentially using agents as lead generators.

“I don’t necessarily know if I believe in a world that there’s apps in the future, but I don’t believe in a world where the apps become data layers of a universal simple UI because when you open your iPhone, there is a reason Airbnb’s app is a different design than DoorDash, Shopify, and Amazon. Our interfaces should be adaptive to the utility and the function that we’re doing. We want to make our app more agentic, and we want to build a lot more user interface controls,” he explained.

He warned that if an agent is to stay within the platform while also offering browsing, messaging, comparison, identity verification, mapping, and other features, either the agent must hand off tasks or the platform must provide a more comprehensive software development kit.

The Need for an AI‑Specific Operating System

Chesky argued that the industry lacks a true operating system built for AI. Current AI tools run on existing OSes like iOS, macOS, and Windows, which are not designed for AI‑level interactions. In his view, AI should operate at the kernel level, enabling agents to interoperate seamlessly.

He illustrated the gap by recalling a conversation with Sam Altman about OpenAI’s attempt to create an app store for ChatGPT.

“If you want to be an app store like the iPhone, you need to have a software developer kit and an operating system like the App Store. Otherwise, none of our functionality is going to exist. And of course, it didn’t work very well,” Chesky said.

Testing the current state, Chesky has used Airbnb through Muse and Instinct and found the experience lacking.

“I’ve used Airbnb on Muse and Instinct. And it doesn’t work very well. I mean, even hotels don’t work that well, but Airbnbs work even worse. So I don’t think we’ve cracked consumer AI,” he added.

He identified two common Silicon Valley misconceptions: that interfaces will disappear in favor of voice or text, and that software will become entirely generative. He maintains that designers can prompt better software than consumers can, and that as long as Apple and Google dominate mobile platforms, apps will remain central, with agents embedded inside them rather than replacing them entirely.

Voice, Efficiency, and the Road Ahead

Looking ahead, Chesky expects voice to become the primary way people interact with computers. Airbnb plans to roll out voice agents this fall to assist with search and customer service, joining a growing wave of voice‑first applications such as Wispr Flow, Willow, and Superwhisper.

On a practical level, Chesky said AI has accelerated feature development and reduced the number of meetings he needs to attend.

“I think AI is reducing communication layers. One of my jobs as CEO is try to get information. And the way I used to get information was having meetings. Now, I can just ask my AI and I don’t need to have as many,” he said.

He envisions a future where the Airbnb app itself becomes a macro agent, with separate agents handling exploration, customer service, and other services, all interoperable through standards such as MCP.

“I think of the Airbnb app as one day becoming an agent, you know, where basically the explore tab has an agent, our customer service has an agent, we’re developing another agent in another service area. Probably eventually there’s going to be a macro Airbnb agent. And that will probably be interoperable with MCP to other agents,” Chesky noted.

For Chesky, the ultimate solution is an operating system built expressly for AI, one that lets agents communicate, share data, and present richer interfaces. Until such a platform emerges, Airbnb will continue to evolve its app, adding agentic capabilities while preserving the browsing and planning experience that travelers value.