AI Chatbots Are Sending Hikers Into Trouble on Real Mountains

Three 20-year-old hikers got lost on Mount Shasta after summiting at 7:30 pm in the dark — and they told rescuers their planning came partly from Google's chatbot, Gemini.

The climbers ran short on food and overestimated their fitness, according to a report from High Country News, part of the Climate Desk collaboration. They should have turned back in late afternoon. Instead, they pushed to the summit at 7:30 pm and had to descend in the dark. By morning, they were lost on a rocky face, and one hiker had banged up his knee.

They admitted later that their lack of preparation came partly from relying on Gemini. Nick Meyers, US Forest Service program manager of the Mount Shasta Wilderness Climbing and Avalanche Center, responded to the call. His team averages a dozen rescues each year.

"We get plenty of people who come unprepared or don't do any research at all. To these guys' credit, they did try," Meyers said. "But I think they kind of turned their brains off."

Is Mount Shasta an isolated case?

No. Meyers said another visitor pitched her tent near a "No Camping" sign and insisted she had a right to stay because AI told her she could. His team also found a man wandering a trailhead with an AI-generated map that put prominent features on the wrong side of a nearby mountain.

The pattern crosses borders. In British Columbia, a helicopter crew rescued a 16-year-old hiker from a ledge in the Widowmaker Arete on the back of Crown Mountain. The teen had used Anthropic's chatbot, Claude, to plan his trip. News stories focused on that fact went viral across Canada and the US — though the hiker later said he had also consulted Google Maps and the popular hiking app AllTrails.

The stakes are real for land managers. Grand Teton National Park in Wyoming has tracked a sharp rise in visitors who report using AI for travel information over the past year. "It went from zero to 26 percent," Park Superintendent Chip Jenkins said. "And I'm guessing that curve is not going to flatten."

Are AI-related accidents actually increasing?

Not yet, according to the available numbers. It remains to be seen whether rising AI use will lead to more backcountry accidents and search and rescues, Meyers said. His office oversees a massive wilderness area in Northern California, and 2026 is shaping up to be an average year for his team.

Wyoming's numbers point the same way. Search and rescue calls have actually dropped compared to 2025: only 312 calls so far in 2026, against a total of 466 in 2025 and 377 in 2024.

The gap between dangerous decisions and actual rescues is where the story gets complicated. Plenty of bad plans end with a scraped knee and a lesson learned, not a helicopter.

Why does this feel familiar?

Some search and rescue veterans say chatbots are just the latest tools giving people new ways to get into trouble. Roughly a decade ago, hikers, backpackers and climbers began sourcing destinations, inspiration and information from Instagram and other social media sites — and sometimes faced deadly consequences. Before that, hikers were led astray by their reliance on Google Maps or smartphone connectivity.

"Over the last two decades, there's been a phenomenal push in the media to get more people outdoors, and consequently we have more people outdoors and recreating," said Jim Moss, a Colorado-based recreation attorney and search and rescue volunteer. It's still unclear, he added, how questionable advice from a robot will influence all those people.

The scale of exposure, though, has changed. A viral Instagram post suggests a place to go. A chatbot hands you a full plan, with confidence, whether or not it knows the terrain.

What makes chatbot advice uniquely risky?

Researchers who study AI and search and rescue point to two specific problems. The first is hallucination — chatbots can present misinformation persuasively as fact. The second is a phenomenon called "cognitive surrender," said Chris Flathmann, a human-centered computing professor at Clemson University in South Carolina who has studied AI and search and rescue.

"When humans work with AI systems, they can get so deep into it they give up thinking critically for themselves," Flathmann explained. When AI pretends to be an expert on an unfamiliar topic, "you can't call BS on that because you don't have the expertise." Once critical thinking stops, he said, it can be hard to restart — especially in a stressful backcountry situation.

Chatbots also tend to tell users what they want to hear, and to do so with authority, regardless of whether their sources are reliable, Flathmann and others said. Claude or ChatGPT might answer a query with extensive directions and advice that sound credible but are actually scraped from Reddit threads and out-of-date YouTube videos — or entirely fabricated.

That's the core difference from earlier technology, in Flathmann's view. TikTok and Instagram offer inspiration. AI offers instructions, right or wrong, plus encouragement.

"AI isn't trying to trick people into going up mountains, but I think it's the natural evolution of what happens when you get a tool like this," he said.

What do experts recommend?

The advice from experts is consistent regardless of the tools people use. Before heading somewhere new, double-check information against reliable, authoritative sources:

• Up-to-date public-land agency maps and websites
• Trail association websites
• Guidebooks
• A real-live guide, national forest official or park ranger

That last option is the one Meyers pushes hardest. His climbing center exists to answer exactly the questions the Mount Shasta hikers put to a chatbot instead.

"If you're going to go into places that will kill you," Meyers said, "you should take your planning seriously."

As AI use keeps climbing — from zero to 26 percent of Grand Teton visitors in a year, by Jenkins' count — land agencies and rescue teams will be watching whether the rescue numbers follow. So far they haven't. The open question is how long that lag holds.