UserGate Warns Blind Trust in AI Is Costing Travelers Their Flights
A Russian cybersecurity vendor named a lack of critical thinking as the top threat facing 32 million local AI users, citing families stranded at boarding gates after trusting chatbot visa advice.
Blindly trusting artificial intelligence has moved from an office nuisance to a tangible travel hazard. According to Svetlana Gazizova, AI security product owner at network security vendor UserGate, the single biggest vulnerability among domestic users is the complete absence of critical verification when interacting with large language models.
To illustrate the problem, Gazizova cited an incident where a family consulted a chatbot regarding visa requirements for a foreign vacation. The model hallucinated an answer, confidently assuring them that no visa was needed. Relying entirely on the generated text, the family bought airline tickets and booked accommodations, only to be turned away at the airport check-in desk for lacking entry documents.
The credulity problem plays out against rapid adoption. Citing market data previously reported by Expert.ru, around 34% of Russians aged 12 and older—roughly 32 million people—use AI tools every month, typically via web browsers, standalone chatbots, and mobile apps.
Beyond travel mishaps, UserGate flagged two other operational risks: sensitive data leaks and underage access. Users routinely paste confidential files and scanned passports directly into consumer LLM prompts without considering how external models store or retrain on the input. Meanwhile, minors can easily query models for prohibited recipes or destructive behavior, skirting basic safety filters.
Gazizova stressed that generative models trained on unvetted public internet text inevitably distort facts. Unless users treat output as an unverified draft rather than an authoritative answer, cognitive offloading will continue to create immediate financial and logistical fallout.
Source: habr
Comments
This is where the magic happens: AI reads your discussion and rewrites the article based on the most interesting comments. Each strong comment adds points to the meter below. Once the meter is full, the article updates live — no page reload needed.