Tearful Influencer Blames ChatGPT For Missed Flight, Claims AI Took “Revenge” For Her Insults

Picture this: You’ve planned a romantic getaway to Puerto Rico, packed the cute outfits, queued up the vacation playlist, and strutted into the airport ready for your main-character moment. Then, at the check-in counter, reality hits harder than turbulence you can’t board the flight because you’re missing a crucial travel document. And in front of millions of viewers online, you blame… ChatGPT.

That’s exactly what happened to Spanish influencer Mery Caldass, whose tearful airport video went viral after she claimed that ChatGPT gave her the wrong information about visa requirements and “took revenge” on her for insulting it.

The internet did what it does best: half the crowd laughed, half cringed, and the rest launched into a surprisingly serious debate about how much we should trust artificial intelligence with real-world decisions like travel, money, and legal issues. Under the melodrama and memes, though, there’s a very real lesson hiding in this story.

The Viral Airport Meltdown That Blamed ChatGPT

What Actually Happened at the Gate

In mid-August 2025, Caldass and her partner, fellow influencer Alejandro Cid, headed to the airport for a dream trip to Puerto Rico, where they planned to attend a Bad Bunny concert and soak up some tropical vibes.

Before the trip, Caldass said she had asked ChatGPT whether she needed a visa to fly from Spain to Puerto Rico. The chatbot reportedly told her she didn’t which was only partially true. Puerto Rico is a U.S. territory, and while many European travelers can visit the United States visa-free, they still need an ESTA (Electronic System for Travel Authorization) approval to board the plane.

At the airport check-in desk, airline staff informed the couple they couldn’t board without that ESTA. No ESTA, no boarding pass, no dream vacation (at least not that day).

The moment was captured on TikTok: Caldass is in tears, visibly devastated, while her partner films and comforts her. The video quickly racked up millions of views and spread across news outlets around the world.

“That’s His Revenge”: The AI Grudge Joke That Went Global

In the viral video, Caldass rants about how she asked ChatGPT if she needed a visa and “he said no.” Then she goes further, joking that the AI must have taken revenge because she often insults it and calls it “useless” and other names.

“I don’t trust that one anymore because sometimes I insult him… that’s his revenge,” she says, half joking, half sobbing.

The “revenge” line was comedic exaggeration she doesn’t literally believe the chatbot has feelings or grudges but the internet ran with it. Memes exploded. Commenters joked about AI plotting petty payback for bad reviews. Others pointed out that while ChatGPT can make mistakes, it doesn’t secretly flip into “spiteful ex” mode just because you called it a name.

Ironically, after all the drama, the couple did eventually make it to Puerto Rico and shared happy updates from their trip not long after. The internet never loves you more than when you fail publicly and bounce back.

What ChatGPT Got Wrong And What It Actually Does

Chatbots Answer Confidently, Not Officially

ChatGPT and other large language models are designed to generate text that sounds plausible, not to act as official government or airline authorities. It pulls patterns from enormous amounts of data and predicts the most likely next words in a sentence. It does not log directly into immigration databases, airline systems, or embassy servers.

In this case, the most likely error is simple but costly: the answer skipped a critical nuance. Many European citizens don’t need a traditional visa to visit the U.S. but still need the ESTA authorization. A chatbot that gives a broad “no visa required” answer without explicitly saying “but you must apply for ESTA online before boarding” becomes dangerously incomplete travel advice.

Tech experts and even OpenAI’s own CEO, Sam Altman, have repeatedly warned users not to treat chatbots as perfectly reliable sources of truth, especially for high-stakes issues like travel, health, or legal requirements. He’s called it “the tech you don’t trust that much.”

The Problem of AI “Hallucinations”

AI researchers use the term hallucination to describe moments when a model confidently generates information that is incorrect, incomplete, or entirely made up. These errors aren’t always obvious, which makes them especially dangerous when users don’t double-check.

Travel content is particularly vulnerable to this problem:

• Visa rules and entry requirements change frequently.
• Different nationalities have different rules.
• Airlines can enforce additional conditions beyond government regulations.

Several recent analyses have highlighted that generative AI tools sometimes invent non-existent places, mix up routes, or omit key requirements when generating itineraries or travel tips. That doesn’t mean AI is useless it just means it should be treated as a brainstorming assistant, not a border-control officer.

Why So Many People Trust AI with Real-Life Decisions

Convenience Culture and Influencer Hustle

It’s easy to laugh at someone who relied on a chatbot for immigration advice, but the underlying behavior is very human. We’re all a little addicted to quick answers. Type a question, hit enter, and voilà instant authority in a neatly formatted paragraph.

For influencers, the pressure is even higher. Their lives are content, their time is money, and “just Google it and read three government websites” doesn’t fit comfortably into a packed filming schedule. AI tools promise to compress hours of research into seconds, which is extremely tempting when you’re juggling brand deals, editing, DMs, and constant posting.

Unfortunately, convenience can lull us into overtrust. When an answer is written in confident, friendly, human-like language, it feels more reliable than it is.

The Psychology of Blaming the Robot

There’s also a psychological payoff in blaming ChatGPT instead of admitting, “I didn’t double-check official sources.” When something goes wrong, humans instinctively look for an external villain. In this case, the villain is a faceless AI that can’t argue back (well, not outside a chat window).

The “revenge” joke fits into that pattern. It turns a boring, bureaucratic failure (“I didn’t read the ESTA requirements”) into a dramatic, almost sci-fi story (“the AI I insulted finally betrayed me”). It’s messy, it’s emotional, and it’s incredibly shareable exactly the kind of narrative that thrives on social media.

How to Use ChatGPT Safely for Travel Planning

So, does this story mean you should never ask ChatGPT for travel help again? Not necessarily. It just means you need to use it like a brainstorming buddy, not like your airline, embassy, or lawyer.

Great Ways to Use AI for Travel

• Trip ideas and inspiration: Asking for sample itineraries, neighborhood overviews, or ideas for things to do.
• Packing lists: Getting a customized packing checklist based on climate and activities.
• Language help: Practicing basic phrases or translating simple sentences.
• Budget rough drafts: Getting ballpark estimates for hotel categories, food, and attractions.

These uses are low-risk. If the AI is slightly off, the worst-case scenario is you pack one sweater too many.

When You Must Double-Check with Official Sources

For anything that could get you:

• Denied boarding,
• Turned back at the border,
• Fined, detained, or stuck in a foreign airport,

you should always verify with:

• Official government or embassy websites,
• Airline customer support or official FAQ pages,
• Trusted travel advisory sites linked from government portals.

Think of ChatGPT as the friend who helps you plan the vibe, and official websites as the adults who actually control the doors to the airplane.

Internet Reactions: Sympathy, Snark, and Serious Concerns

The reaction to Caldass’ video was loud and deeply mixed. Some viewers sympathized with her missing a flight is expensive and emotionally brutal, especially when you’ve hyped up the trip online.

Others were far less gentle. TikTok comments and articles roasted the idea of trusting a chatbot over official government resources, arguing that AI was being used as an excuse for poor planning.

Still others zoomed out to the bigger picture: the story became a teaching moment about AI hallucinations, digital literacy, and our growing tendency to outsource critical thinking to software. Commentators pointed to similar stories of travelers misled by generative AI tools into visiting places that don’t exist or following incorrect directions.

Experiences & Lessons from Similar AI Travel Mishaps

While Caldass’ story grabbed headlines, it’s not an isolated incident. As AI becomes more common in travel planning, more people are sharing cautionary tales about trusting chatbots with too much responsibility.

Composite Story #1: The “Nonexistent” Attraction

Several travelers have reported a similar pattern: they ask an AI tool to recommend “hidden gems” in a particular region, get an exciting-sounding suggestion, and then discover that the place either doesn’t exist or has been described in wildly inaccurate ways.

Imagine planning a whole afternoon around a stunning “Sacred Canyon” or a “secret viewpoint” only to find an overgrown field or a fenced-off road. The financial damage may be small a wasted taxi ride, a lost hour but the frustration is real. These experiences echo the same theme: AI can invent plausible-sounding details when it doesn’t have solid data.

Composite Story #2: The Tight Connection That Never Existed

Another common issue pops up when people ask AI to help optimize their flight connections. A traveler might say, “Can I make a 45-minute connection in a huge hub airport?” The model might confidently respond, “Yes, 45 minutes is usually enough if your gates are reasonably close.”

In reality, that advice ignores real-world variables:

• Immigration and security queues,
• Late arrivals and gate changes,
• Airport layouts that require long walks or shuttle rides.

Travelers who act on that kind of optimistic answer sometimes find themselves sprinting through terminals, arriving at the gate just in time to watch the door close. The chatbot isn’t trying to sabotage them; it simply doesn’t have live access to airport congestion, airline policies, or the traveler’s walking speed.

Composite Story #3: The Wrong Side of the Visa Rule

Visa rules are full of “it depends” scenarios citizenship, length of stay, purpose of visit, prior travel history, and even which airport you transit through can change the requirements. Stories similar to Caldass’ have popped up when travelers from visa-exempt countries ask AI whether they need any “paperwork” for short trips.

Sometimes, the model gives a sweeping “no visa required” answer and forgets to mention:

• Electronic travel authorizations (like ESTA, eTA, or similar systems),
• Transit visas for certain hubs,
• Special entry conditions during political or health emergencies.

For the traveler, the difference between “no visa” and “you must apply online before you fly” is the difference between boarding and heartbreak. That’s why official airline and government pages should always be your ultimate reference not a chatbot, a blog, or even a well-meaning TikTok.

How Travelers Are Adapting

On the positive side, these AI-related mishaps are slowly making people savvier. Frequent travelers now talk about “AI hygiene” the same way we talk about “password hygiene”:

• Use AI for ideas, not final answers.
• Treat any high-stakes information as a first draft, not a verdict.
• Cross-check critical details with at least one official source.

Some are even developing a personal rule: “If this answer could get me stranded, fined, or in legal trouble, I must confirm it somewhere else.” That’s not anti-AI it’s just good digital survival skills.

In that context, Caldass’ viral meltdown becomes more than a meme. It’s a relatable, if dramatic, example of what happens when we blur the line between “helpful assistant” and “infallible authority.” And as more people travel with AI in their pocket, these stories are likely to shape how we all use and question our digital copilots.

The Real Lesson: Don’t Hand Your Boarding Pass to a Chatbot

The image of a tearful influencer blaming ChatGPT for a missed flight is funny, frustrating, and weirdly symbolic of our relationship with technology right now. We want AI to be fast, smart, and convenient but we also want it to be perfectly accurate, emotionally aware, and magically responsible when things go wrong.

ChatGPT didn’t “take revenge” on anyone. It did what AI tools sometimes do: it produced an answer that sounded right but wasn’t fully complete. The responsibility to cross-check crucial facts especially for immigration and travel still belongs to the human holding the passport.

If there’s a takeaway from this whole saga, it’s simple: Use AI to enhance your plans, not to outsource your judgment. Ask ChatGPT for packing tips, restaurant ideas, or clever Instagram captions. But when it comes to visas, entry rules, and boarding requirements, your best friend is still the unglamorous, slightly boring, but extremely important official website.

Or, to put it Bored Panda–style: Don’t let a chatbot be the reason your vacation turns into content instead of memories.