ChatGPT Scales Back In-platform Bookings, Affecting Hotels

OpenAI announced a reduction in the scope of planned in-platform purchasing for ChatGPT, refocusing the experience on product discovery rather than completing transactions, HospitalityNet reports. Hotel News Resource published an analysis by Apollos Gause on May 6, 2026, outlining implications for hotels and guest-facing chatbots. Skift reports that shares of Expedia and Booking Holdings rose in response to OpenAI's retreat from checkout features, and Skift cites a joint Skift Research and McKinsey figure showing the share of travelers using tools like ChatGPT "extensively" for trip planning rose 124% year over year, from 13% to 30%.

Companies and websites that rely on search and referral traffic will increasingly compete for AI citations and snippet placement. Industry commentary recommends focusing on structured, semantically rich content, site-level trust signals, and clear booking funnel links so AI systems can cite brand pages as the completion endpoint. For practitioners, that implies investment in reliable metadata, schema markup, canonicalized availability feeds, and content models that expose context AI needs to evaluate options accurately.

Reporting frames the OpenAI change as part of a broader tension between discovery (AI-driven research and itineraries) and transactions (checkouts on platforms). Skift's coverage stresses that even without in-platform checkout, AI tools already reshape the top of the funnel by influencing traveler preferences and narrowing choices before users reach OTA or hotel sites. Hotel News Resource and HospitalityNet position this shift as an opportunity for hotels to recapture direct-booking traffic via optimized content.

• •Adoption of Generative Engine Optimization (GEO) practices across hotel websites and CMS platforms, including semantic content models and structured data.
• •Integration points that let hotels expose live rates and availability in machine-readable feeds that AI systems can cite without hosting checkout.
• •Competitive responses from OTAs and search incumbents; Skift highlights investor reaction in public markets to the checkout walkback.

Editorial analysis: Observed patterns from comparable platform walkbacks show that when discovery and transaction layers separate, organizations prioritizing clean API endpoints, canonical booking URIs, and high-quality content capture disproportionate referral conversions. Engineering and product teams should treat AI-driven discovery as a new upstream channel and instrument referral paths to measure AI-origin traffic quality and conversion.

What has been reported focuses on OpenAI's product scope change and industry reaction; the company has not, in the cited coverage, provided detailed public technical documentation about how ChatGPT will source or cite specific booking endpoints. Sources used here include HospitalityNet, Hotel News Resource, and Skift.