AI agents reshape travel, but frauds can surge fast, tells DataDome

Agentic commerce: a new security paradigm

Traditional bot-detection systems, long used by e-commerce players, are no longer sufficient. Until now, the goal was to distinguish human traffic from automated visits and block the latter. With the rise of AI agents, the challenge has evolved: companies must now verify visitor identity and assess intent in real time.

Businesses need to differentiate between three types of agents: legitimate AI agents acting on behalf of real consumers; malicious bots designed to scrape pricing data; and compromised or impersonated agents that misuse legitimate systems by hijacking real user identities. The focus is no longer just on bot detection, but on building a trust model based on intent.

Travel booking relies on complex processes and constantly changing data such as dynamic pricing, fluctuating availability, and coordination across multiple services. Faced with this complexity, more consumers are turning to AI agents to save time and optimize decisions. However, this trend also increases exposure to attacks. More than 90% of travel sites are reportedly not fully protected against even basic automated threats, according to DataDome.

Key threats facing the industry

Agent takeover and account fraud
Fraudsters can compromise an agent’s credentials or hijack a session to make unauthorized bookings, steal loyalty points, and access sensitive data such as passports and payment details. Malicious bots can carry out seemingly legitimate transactions, bypassing traditional systems designed to flag unusual human behavior.

Inventory security and price manipulation
AI agents can extract large-scale data on pricing, availability, and inventory. Some competitors use this intelligence to adjust pricing in real time, while malicious actors may attempt to disrupt the market by blocking inventory or creating artificial demand.

Loyalty program fraud
Travel loyalty programs, valued at $328.02 billion in 2024 according to Market Research Future (MRFR), are prime targets for automated fraud. Bots use stolen credentials to perform credential stuffing, then redeem points for flights or hotels, or resell them on the dark web. Beyond financial losses, this erodes trust, increases customer service costs, and drives attrition among high-value travelers.

Why travel brands must become “AI-friendly”

By enabling secure access to their platforms -capable of distinguishing legitimate from malicious agents in real time, travel companies can boost visibility, reliability, and competitiveness:

Greater visibility: facilitate access for legitimate AI agents to strengthen presence in search and recommendations

Improved customer experience: streamline journeys and protect conversions by reducing friction for real users

Protected margins: block price scraping and inventory hoarding to maintain competitiveness

Stronger partnerships: ensure clean traffic and secure systems to sustain ecosystem trust

Safeguarded reputation: protect loyalty accounts and prevent fraudulent bookings to maintain traveler confidence

For DataDome, travel brands that successfully adapt will be well positioned to capitalize on agentic commerce, while safeguarding their infrastructure and customer trust.