IHG Launches Conversational AI Search Beta in the U.S.

The launch connects a natural-language interface to IHG's operational hotel catalog rather than stopping at trip inspiration. According to the company announcement, recommendations use IHG's verified property information and guest reviews, including imagery and amenity details.

That grounding is the core technical requirement. A hotel recommendation can be fluent but still fail the customer if room availability, rates, property attributes or loyalty redemption options are stale. IHG's design brings those systems into the discovery flow and then carries the user toward booking.

The company also plans to introduce an AI-optimized content platform over the next several months. Hotels will be able to add richer material such as floor plans, videos and updated images so their distinctive attributes can be represented in AI-powered discovery on IHG and third-party channels.

Skift described IHG as the third major hotel group to formally launch conversational search, after Hilton's AI Planner in March and Marriott's Ask Bonvoy in June. Its reporting also says the U.S. beta is being exposed gradually, so not every user will necessarily see the AI option at the same time.

The points-and-cash path is a meaningful distinction for a loyalty-led hotel business. It tests whether a conversational interface can handle both discovery intent and the constraints of rewards inventory without forcing users back through a separate search flow.

IHG's broader data and reservation work matters here. Skift reports that the company has moved guest data to Google Cloud and is migrating to an Amadeus-built cloud reservation system. Those projects are separate from the interface, but reliable content, customer and inventory APIs determine what the assistant can safely surface.

IHG has not published engagement, recommendation-quality or booking-conversion results for the beta. Skift reports that major hotel groups generally have not released comparable performance data for their conversational search products.

For data and product teams, the useful measures will include search-to-booking conversion, abandonment, rate and inventory accuracy, recommendation relevance, loyalty redemption behavior and performance across trip types. Until those results are available, the launch demonstrates production integration and distribution strategy, not a proven lift in bookings.