What Is the Role of Artificial Intelligence in the Outdoors?
Artificial intelligence is rapidly expanding into the outdoor community. Its impact stretches towards how adventurers plan, prepare and experience the natural world. Increasingly, AI is powering trail recommendations, navigation, wildlife monitoring and even personalized gear suggestions. But what does this mean for the future of adventuring? What does it mean for outdoor brands and the community? For the environment? Is AI redefining our connection to the outdoors?
Bike technology is always evolving. Whether it’s the materials the bike is made of, the design of the frame or the ins and outs of suspension and shifting, companies are always driving towards the next big innovation.
But some things never really change. Bikes always have one seat, two wheels and a single set of handlebars. But not the bike that Recreational Equipment Incorporated, better known as REI, posted in an Instagram ad earlier in June. That bike had two sets of handlebars. Flat bars in the front and dropbars in the back. But this wasn’t a bold new innovation in bike design; it was an AI hallucination. And it didn’t take long for people to notice.
“REI using AI slop now. So much for caring about the environment,” read one Reddit post, just one of the hundreds of comments about the ad. REI quickly pulled the ad, pointing out that it was Meta’s AI advertising tools that had altered the image.
“Meta auto-enrolled us in an AI personalization tool that produced an inaccurate and inappropriate alteration of a vendor-provided image in some of our ads,” a REI spokesperson told Business Insider. “While a two-handled bike might be interesting, it is not something you will find in our assortment.”
It was just one ad, but it’s a sign of artificial intelligence’s growing presence in our lives. And it’s raising questions about what that means for the outdoor community.
Should You Use AI for Trip Planning?
In 2023, reporters at The Atlantic published a database called Book3 that listed the nearly 200,000 titles of books used, without permission, to train generative AI systems. Among them, were two British Columbia hiking guidebooks authored by Taryn Eaton.
“It’s copyrighted material; it’s been stolen,” Eaton, the Squamish-based author of Backpacking in Southwestern British Columbia and Backpacking on Vancouver Island, explained. “I know that my website is being used to train AI. If I put my own queries into ChatGPT or Gemini, it sometimes gives me back my own information.
And Eaton isn’t the only one. Dozens of outdoor book titles, from guidebooks to adventure narratives, were recorded in the Book3 database. Online articles, like this one, are also notoriously hard to protect from crawler bots, automated programs that gather online data to train AI models.
“The biggest frustration for me is that, as a full-time outdoor writer, this is my living and my livelihood,” Eaton said. “It makes it very difficult to do this for a living if my writing is being stolen and used in ways that I didn’t consent to.”
But AI isn’t just stealing information from outdoor writers and guidebooks and storing it; it’s using it. In some cases, to give people advice on where to hike, camp and backpack.
Using AI for travel planning isn’t new. But its integration with outdoor navigation tools is relatively recent. Earlier this year, AllTrails announced that it had added AI “assistants like ChatGPT and Claude to help you discover and learn about trails”. The features, they claim, will let users get trail information and conditions more easily.
Outside Interactive Inc, the new parent company of the outdoor magazine-turned-tech empire Outside, has had its version for even longer. Called Scout, it was launched as an “adventure concierge” in 2024.
“Are you looking for a more comfortable pair of ski boots? Expert advice on buying your first e-bike? A guide to the Grand Canyon from someone who’s hiked hundreds of miles there? Scout can provide detailed responses to your outdoor queries in just seconds,” the press release announcing Scout claimed.
But is the information these tools provide any good? For some people, AI might make finding trail information easier, but the information can also be out of date or just wrong.
In 2025, an editor at Backpacker Magazine used AI to plan a hike in the North Cascades. The tool helped them narrow down a trail, but “the bot couldn’t tell me anything about the current weather report, road conditions or trail closures.”
Taryn Eaton had a similar experience when she tested out an AI trip planner.
“I asked ChatGPT to help me plan a hike in the Stein Valley, and it gave me an itinerary that was incomplete,” she said. The trip AI suggested was neither a loop nor an out-and-back, it just ended in the middle of the wilderness, Eaton explained. It also included side trips that weren’t along the route, and sections of the trail that weren’t accessible.
“Even though it cited BC Parks and several other sources, it didn’t flag that BC Parks has said that some of the suspension bridges and cable cars on the trail are broken,” she said. “[The AI didn’t know] that the trail is actually closed in that middle section.”
The problem, Eaton explains, is that AI models don’t have the kind of knowledge that’s earned through miles on the trail.
“I can tell people things about those trails that they might not get from an AI chatbot, or from AllTrails,” she said. “I can tell them about where the most responsible place to go to the bathroom is, if they can’t find a toilet. I can tell them where to filter water or the best time of year to see certain plants or animals. That kind of information is not readily available without experience.”
Ironically, AI actually seems to agree with Eaton. When Mikaela Osler, a guide with Skurka Adventures, asked the chatbot Claude if it was safe to let AI plan a backpacking trip, it said no.
“AI does not ‘know’ wilderness terrain in the way a guide or ranger does,” the chatbot explained. “It predicts plausible language patterns based on training data. Sometimes those patterns correspond to reality. Sometimes they do not.”
What Does AI Mean for Trails and the Environment?
But even when AI’s trail information is accurate, it may still contribute to other challenges, such as overcrowding. When I was working on this story, I did an experiment, asking AI to suggest some local hikes near popular outdoor towns in British Columbia.
Nearly every top suggestion was one of the region’s busiest trails. Where I live near Chilliwack, it suggested Lindeman Lake, probably the busiest hiking trail in the Fraser Valley. In Squamish, it suggested the Stawamus Chief. In Pemberton, the number one suggestion was Joffre Lakes, one of British Columbia’s busiest trails.
“If you had asked me a few years ago, I would have said I don’t think that we could put any more pressure on popular places than social media is already doing,” said Eaton. But AI is doing just that, directing people to trails already experiencing overuse and overcrowding.
“It’s a feedback loop between AllTrails, social media and AI,” said Eaton. “They’re all just feeding each other the same 10 popular choices in each location.”
And when it comes to the environmental impacts of AI, trail overuse is just the tip of the iceberg. Running AI systems requires an inordinate amount of computing power. To get that, companies are building data centres around the world.
According to a 2025 United Nations Environment Program report, these centres had four main environmental costs. First, a massive amount of raw material is needed to create the computers. This raw material contains many critical minerals, the mining of which is already threatening both wild places and public recreational destinations, such as the Seven Summits trail outside of Rossland, British Columbia.
Second, the UNEP report explained, “Data centres produce electronic waste, which often contains hazardous substances, like mercury and lead.”
Third, data centres use a lot of water. While each query doesn’t use much water—estimates range from a few millilitres to a bottle’s worth—the sheer volume of queries adds up quickly. Large data centres use upwards of 5 million gallons of water per day, while worldwide, it’s estimated that AI infrastructure could consume six times as much water as all of Denmark’s six million residents.
And then there’s the question of power. A single ChatGPT query uses 10 times as much energy as a Google search, and as data centres expand, so too do proposals for new power plants and hydroelectric dams. Like critical-minerals mines, these dams could put beloved outdoor places at risk, including the Homathko River, where the paddling advocacy group BC Whitewater is raising concerns that the province is considering a major new hydro project.
Are Outdoor Brands Using AI?
Shortly before REI posted that unfortunate ad, Canadian Prime Minister Mark Carney made a much more consequential announcement. His Canadian AI strategy, dubbed “AI for All,” laid out ambitious plans to expand AI infrastructure and use it across the country. With billions of dollars in investment, the goal is “increase AI adoption from just over 12% to 60% by 2034” in Canada.
There are, of course, legitimate uses for artificial intelligence. Taryn Eaton said she uses it for grammar. As a writer, I’ve used it to help transcribe interviews, a process that, when I started working as a journalist, would take hours of my day. But this level of scaling raises serious questions about the human and environmental costs, questions that the government has yet to answer.
It also raises ethical questions, such as whether companies, including outdoor brands, are using AI in their advertising. After the backlash to their ad, REI distanced itself from the photo. They pointed out that Meta had altered the ad with AI and that their original ad was a real photo provided by Van Rysel, the bike brand pictured in the shot.
AI is a complicated tool. And it’s one that’s increasingly showing up in, and creating challenges for, both the outdoor community and the places we love and rely on. It’s not a clear black-and-white situation, and since this story won’t be going anywhere, we’ll keep tracking it as best as we can to keep you informed. When we do, you can be sure it will be a person writing the stories and taking the photos.
