Planning a holiday should be an exciting escape, a chance to unwind and create beautiful memories. Today, many of us turn to Tripadvisor to help us choose the perfect destination, often relying on the platform’s new AI-generated summaries perched conveniently at the top of hotel pages. These quick, easy-to-read snippets promise to take the hassle out of trip planning by instantly condensing hundreds of reviews. However, a deeply concerning investigation by the UK consumer champion Which? has revealed that these digital shortcuts are doing far more than saving us time—they are glossing over severe health hazards, stomach-churning hygiene failures, and incredibly frightening safety issues, leaving unsuspecting travelers in harm’s way.
The most alarming example of this disconnect is found at the five-star, all-inclusive Riu Palace Santa Maria in Cape Verde. While Tripadvisor’s AI enthusiastically pitches the resort as a “spotless” favorite with “rave reviews” for its dining, the reality shared by actual guests is nothing short of a nightmare. Holidaymakers reported rampant food poisoning, raw chicken, and birds and flies feasting on buffet food, with one terrified guest spotting dead mice near the seating area. Shockingly, the resort is currently the subject of a massive group legal action involving over 400 ill travelers and seven reported deaths since 2023. Yet, when tested, Tripadvisor’s AI chat assistant, Ollie, assured users that food poisoning was “quite unlikely” and praised the hotel’s “high hygiene standards.”
Unfortunately, this is not an isolated glitch; it represents a systemic pattern of AI downplaying major hazards. At the Garza Blanca resort in Cancun, where a wedding party and multiple other guests fell violently ill, the AI blindly promised “immaculate cleanliness.” Even worse was the summary for the Occidental Caribe in the Dominican Republic. While real people described a harrowing lack of running water, rooms smelling of raw sewage, and widespread mold, the AI chose to highlight the hotel’s “abundant amenities,” reduced the severe structural issues to minor “maintenance issues,” and called the cleanliness merely “inconsistent.”
Perhaps most disturbing is how the AI handles personal safety and trauma. At the Kaia Coracesium on Turkey’s Antalya coast, multiple female guests left frantic warnings detailing persistent sexual harassment by male hotel staff, including workers following young girls to their rooms to demand social media details. Instead of flagging this vital safety warning, Tripadvisor’s AI sanitized the feedback. It characterized the hotel service as generally “friendly,” dismissing the terrifying experiences of these women and families as mere “lapses in service.” This cold, corporate translation reveals a dangerous inability to distinguish between a cold cup of coffee and a serious threat to personal safety.
When confronted with these findings, Tripadvisor defended its technology, explaining that its algorithms digest reviews from the past year to highlight the most common themes. They emphasized that these summaries are meant to be snapshots, not a replacement for reading actual reviews, and argued that travelers do not need to “blindly trust” the AI. However, critics are quick to point out the hypocrisy of this defense. As Rory Boland, the editor of Which? Travel, passionately argued, if these summaries are not meant to be trusted, Tripadvisor should not be placing them at the very top of their pages where they inevitably capture the most attention. By burying life-threatening risks under fluffy, automated praise, the platform is putting profits and convenience ahead of human lives.
For those planning their next getaway, the message is clear: do not let artificial intelligence do your homework. Other search engines, like Google, have proven much better at accurately capturing critical warnings, with Google’s AI actively flagging “outbreaks of illness” and “potential for disease” at the very same properties Tripadvisor praised. Until these algorithms are trained to value human safety over marketing jargon, the best way to protect your family is to scroll past the AI summaries entirely. Take a few extra minutes to read the real, unfiltered stories of fellow travelers—especially the one-star reviews—to ensure your dream vacation doesn’t turn into a tragedy.
