TL;DR

Major travel companies including Expedia, Google, Kayak, and Priceline are developing agentic AI tools capable of autonomously booking travel. Whilst 80% of travel executives plan to offer such tools within five years, just 2% of consumers are ready to grant AI booking autonomy.

Beyond Search to Action

Approximately 30% of US travellers already use AI to compare flights, hotels, and rental cars. But agentic AI represents a fundamental shift: rather than simply summarising information, these tools can carry out tasks independently.

“Instead of manually running every search or comparison yourself, agentic A.I. can do that work in the background and come back with the most relevant options,” explains Shilpa Ranganathan, Chief Product Officer at Expedia Group.

The technology enables far more sophisticated searches than current generative AI. When comparing hotels, an agentic tool can pinpoint multiple properties’ rates for specific dates, check availability, and compile everything for easy comparison—rather than simply suggesting options and providing links.

The Trust Gap

The technology exists to automate purchases entirely, but consumer readiness lags significantly behind. According to Skift’s State of Travel 2025 report, just 2% of travellers are willing to give AI autonomy to book or modify plans. A Booking.com survey found only 12% comfortable with AI making independent decisions.

Privacy concerns compound the hesitation. More than half of consumers worry about how companies use their data—a significant hurdle when agentic AI requires credit card information to function autonomously.

“We don’t yet have a view on how much autonomy customers may want,” acknowledges Cobus Kok, Vice President for Product and AI at Priceline.

Guarding Against Hallucinations

AI’s tendency to fabricate information poses particular risks in travel, where inaccurate hotel, flight, or visa information could have serious consequences.

“In travel, accuracy is everything,” says Ranganathan. “If A.I. suggests the wrong hotel, flight or visa requirement, it can erode trust instantly.”

To mitigate risks, Expedia’s AI draws from verified booking data rather than the broader internet, whilst other companies emphasise rigorous testing and customer feedback mechanisms.

Looking Forward

Steve Hafner, Kayak’s CEO, describes the development as “a multiyear journey,” acknowledging that consumer habits will need time to adjust. For businesses, agentic AI promises significant efficiency gains—but building the trust necessary for adoption remains the industry’s primary challenge.

Source: The New York Times