Imagine going hiking trough Peru and not triple checking the information. Even legit source can become incorrect quickly when it comes to hiking trails. Besides not checking the information they went without a guide, yeah these people were definitely noobs who came unprepared. Bet they didn’t even bring a satellite phone. This is how people die. Like two girls from my country went hiking in Panama, they went without a guide because the guide they booked didn’t arrive on time. They never returned. Search and rescue only found a camera and some parts of their bodies.
this post was submitted on 05 Oct 2025
92 points (97.9% liked)
Fuck AI
4230 readers
618 users here now
"We did it, Patrick! We made a technological breakthrough!"
A place for all those who loathe AI to discuss things, post articles, and ridicule the AI hype. Proud supporter of working people. And proud booer of SXSW 2024.
founded 2 years ago
MODERATORS
When I was on my cross-Canada trip with SO in 2024, there was a time when we were on a gondola lift with a bunch of the younger generation. They were planning a trip to Ottawa (Ottawa being sort of my stomping grounds of over 20 years). They were asking ChatGPT for things to visit and then commenting on them out loud. Which allowed us to hear that well over half the "interesting sights" they were planning on seeing in Ottawa didn't exist. Some of them were locations in Montreal (Concordia University campus) or elsewhere in Ontario (UWO campus). But the rest just didn't exist anywhere at all as far as I knew.
Even worse, the ones ChatGPT uttered that did exist they put in the wrong sections of the city. Kind of like going to Queens to see Broadway in New York.
I wonder how many times an AI needs to fuck up before people just lose confidence in it?
I wonder how many just blindly believe the bots without fact checking the destinations, cuz wasn't there a guy who asked the llms if he needs a visa to go to Peru iirc and it said no, so he he flew to Peru and couldn't go any further because he didn't have a visa?
I wonder how many just blindly believe the bots
To be fair, the bots are being explicitly marketed as believable. The customers are being lied to. I personally try to have sympathy for them and direct my frustration at the legislators not enforcing false advertising laws.
Who wouldn't want to visit a Winery in the middle of the Atlantic Ocean? Or go to a world class water park in the Sahara Dessert? I heard there was a new Super McDonalds opening on the Moon on Halloween. What a time to be alive.
probably the only LLM I've liked for searches is Kagi's, because it tries to not answer your question. it figures out search terms for it, then does the search, summarizes or approximates an answer, them gives citation links so you can check it
liking LLMs? That's whack.
FAFO
I think it's important not to victim blame here. These people were lied to, by the bots, and by the companies that say the bots are trustworthy. Their government that permits the false advertising is failing them.
People have a responsibility to themselves. One of the absolute first things I did when I heard about ChatGPT and the ever-increasing coterie of its imitators, was I tested them. I had them talk about things I know and counted the errors and flat-out hallucinated fictions.
Then I said to myself, "you know, they're going to be just as full of shit on things I don't know".
I saw all the lies. I saw all the advertising. I saw the same thing everybody else saw. But I saw it all and then actually tested it.
If people are willing to just believe professional liars—and make no mistake, that's what advertisers are!—without bothering to do a minute's checking, then sorry, that's entirely on them.