DÜSSELDORF, Germany — Trivago is making a calculated bet that artificial intelligence will reshape hotel search without eliminating the traveler’s instinct to compare prices.

In an interview highlighted by CNN, chief executive Johannes Thomas outlined two sides of the same strategy: apply AI to the consumer-facing search experience while embedding it throughout the company’s internal operations.

The first effort is designed to protect Trivago’s position as travelers migrate from conventional search engines toward conversational services. The second is intended to make a company of roughly 600 core employees operate with the reach and speed of a much larger organization.

Trivago has not announced a universal tool that independently compares live hotel rates inside every major AI assistant. Its current technology largely uses AI within Trivago’s own hotel-search environment. The broader question—how Trivago’s price-comparison capabilities might appear within platforms such as ChatGPT, Gemini or Claude—remains part of an emerging industry contest.

Turning a Travel Request Into a Hotel Search

Trivago introduced its beta Smart AI Search in November 2024. Developed with Google Cloud and its Vertex AI Search technology, the feature allows travelers to describe what they want conversationally: a rustic luxury hotel near the Canadian Rockies, for example, or a family property with a pool, parking and easy access to hiking.

The system interprets the request, identifies relevant properties and returns Trivago’s familiar visual results. It therefore adds natural-language understanding to metasearch rather than replacing the underlying comparison marketplace. Travelers can still assess rates supplied by online agencies, hotel groups and, in some cases, hotels themselves.

The company has since expanded its use of generative AI. AI-generated hotel highlights and review summaries now condense large amounts of property information into more readable assessments. In August, Trivago said it had improved context-aware highlights for more than 500,000 hotels, allowing results to reflect more closely what an individual traveler is seeking.

This is the most accurate interpretation of the first CNN headline: Trivago is adapting hotel-rate comparison to an AI-led search environment. It is not yet evidence that travelers can enter any outside chatbot and receive a complete, Trivago-powered comparison with guaranteed live availability.

The Missing Link Is Live Inventory

Conversational recommendations are relatively easy; reliable booking data is harder. Hotel prices can change within minutes, and apparently identical offers may differ in taxes, breakfast, cancellation rights, room type or membership conditions.

That is why the industry is now building infrastructure that allows AI assistants to retrieve current rates rather than relying on previously indexed webpages. In April, Australian hotel-technology company SiteMinder announced connections intended to place participating hotels, live prices and availability within AI channels. DirectBooker was named as its first partner in this effort. SiteMinder’s Demand Plus product already distributes hotels through established services including Google, Trivago and Tripadvisor, but SiteMinder—not Trivago—is developing this particular cross-platform connection.

The distinction matters. Trivago remains principally a hotel metasearch company: it organizes offers, helps travelers compare them and normally sends the customer to a booking provider. Its expanding Book & Go service creates a more consistent booking path, supported by technology acquired through Holisto. Trivago said Book & Go’s share of bookings roughly tripled from a year earlier and that Expedia had joined as a supply partner. But this is still different from completing an autonomous purchase inside a general-purpose chatbot.

AI Behind the Screen

In the first seven months of 2026, the company spent more than five times as much on internal AI tools and computing tokens as it had during all of 2025. Thomas said the company does not operate a single centralized AI budget: spending is distributed across more than 100 teams, including engineering, finance, legal, marketing and human resources. The company has not disclosed the absolute amount.

According to Trivago’s internal survey, 93 percent of employees were using AI daily, up from 63 percent a year earlier. Employees estimated that the tools saved them an average of 55 minutes a day, compared with 36 minutes the previous year, while 86 percent said AI produced a measurable improvement in their output.

Those figures suggest unusually broad adoption, but they remain company-reported and largely self-assessed. They demonstrate employee enthusiasm more clearly than they prove a direct financial return.

Trivago’s internal system includes its own Copilot, alongside approved external services and specialized agents for coding, analysis, contract review, marketing and content production. Thomas has described the goal as becoming an “AI-native” company in which agents perform more execution while employees concentrate on judgment, direction and creative quality.

That language points toward a redesign of work, although Trivago has not presented the program simply as a plan to replace employees. Its stated ambition is to give approximately 600 people the productive impact of 6,000—a striking claim that will ultimately be judged by product quality and operating margins, not minutes reportedly saved.

A Defensive Strategy Against Google—and the Chatbots

For Trivago, AI is not only an opportunity. It threatens to relocate the starting point of travel planning.

Google’s AI Mode can already assemble itineraries using real-time information from Search, Maps, flights and hotels. Booking.com has developed conversational planning and support tools, while Expedia is pursuing AI-powered discovery and booking. The competitive risk is that travelers may receive a satisfactory hotel recommendation before ever reaching a traditional metasearch site.

For now, that transition remains limited. Expedia’s April survey of more than 5,700 adults in the United States, Britain and India found that only 8 percent used AI assistants for trip planning, while 68 percent preferred to complete bookings with an established travel brand. Concerns included control, privacy and accountability when a reservation goes wrong.

Thomas has similarly said that traffic arriving from AI platforms remains small, though it is growing and converts at comparatively high rates. His argument is that general-purpose chatbots may excel at inspiration but still struggle with the visual, iterative and detail-heavy process of selecting a hotel.

The relationship with Google is particularly complicated. Trivago uses Google Cloud technology and participates in Google’s hotel-advertising ecosystem, yet in May it sued Google in Hamburg, alleging that the search company had systematically favored its own hotel-comparison service between 2014 and 2025. Google’s liability has not been established, and Trivago’s regulatory filing describes the proceedings as being at an early and uncertain stage.

A German Company Under European Rules

Trivago’s transformation is also unfolding as the European Union’s AI Act becomes broadly applicable. Transparency requirements that took effect on Aug. 2, 2026, are intended to help consumers recognize when they are interacting with AI or viewing AI-generated material. Earlier provisions already required organizations to promote appropriate AI literacy among employees.

For a Düsseldorf-based company generating hotel summaries and deploying AI across legal, marketing and personnel functions, transparency, data protection and human oversight will therefore be as important as speed.

Trivago’s early financial evidence is encouraging but does not isolate AI’s contribution. Second-quarter revenue rose 21 percent to €168.4 million, product conversion had increased 64 percent since the second quarter of 2023, and the company recorded positive adjusted earnings before interest, taxes, depreciation and amortization. Those gains also reflect renewed brand advertising, product simplification, membership offers and changes to its commercial model.

The larger test is whether Trivago can preserve its role as an independent comparison layer when AI assistants begin controlling more of the journey from inspiration to purchase. Its strategy is becoming clear: make Trivago more conversational for travelers, more automated for employees and more valuable as a source of trusted comparisons.

The risk is equally clear. If the major AI platforms can obtain live hotel inventory directly and present convincing comparisons of their own, Trivago may find itself supplying intelligence to a marketplace whose customer relationship is controlled elsewhere. For now, the company is trying to ensure that the arrival of AI does not remove the middleman—but makes that middleman faster, leaner and harder to bypass.

Sources: TRIVAGO, PhocusWire, EU Commission, Expedia