Detect Deepfakesby Resemble AI
Deepfake case study · Image

Help me convince my friend organizing a trip to Amsterdam

Travelers booking accommodation in Amsterdam encountered a fraudulent listing on Booking.com featuring AI-generated images and inconsistent property details

Reported date
Sep 30, 2026
Target
Travelers booking accommodation in Amsterdam

The exact incident date was not established. This entry is dated by its source report.

Updated Sep 30, 2026 · 1 min read

A fraudulent accommodation listing on Booking.com recently targeted travelers seeking lodging in Amsterdam by utilizing AI-generated imagery to misrepresent a property. The listing attempted to lure potential guests with unrealistic pricing and luxury interior photos that failed to align with the actual physical characteristics of the building.

What happened

The scam involved a coordinated effort to deceive users through several discrepancies between the provided listing information and reality. The interior photographs appeared to be AI-generated, showing a spacious layout that contradicted the actual size of the property recorded in the land registry. While the listing claimed the unit was 105 square meters with four bedrooms and two living rooms, official records indicated the space was significantly smaller, measuring between 53 and 64 square meters.

Further investigation revealed multiple red flags regarding the host and the property's authenticity. The host's description claimed to be a 38-year-old Dutch woman, yet the provided name was "Harry." Additionally, the interior photos did not match the exterior view of the building, and the listing contained conflicting information regarding the number of bedrooms. The listing also featured only two perfect reviews, both submitted on the same day from users in Spain, which raised suspicions about the legitimacy of the feedback. Travelers are advised to verify property addresses using tools like Google Street View and to cross-reference interior photos with exterior building features to identify potential inconsistencies.

Sources