is a senior science reporter covering energy and the environment with more than a decade of experience. She is also the host of Hell or High Water: When Disaster Hits Home , a podcast from Vox Media and Audible Originals.
A popular ski resort town in Colorado is adopting a new AI Smart City Solution from Hewlett Packard Enterprise (HPE) to help it better detect wildfires, as well as update a range of other city services.
Vail is expanding its firefighting toolbox as hotter, more arid weather with climate change raises wildfire risk in the western US. Colorado has suffered 11 of the 20 largest fires in state history just within the last five years. Researchers and first responders are increasingly exploring ways in which new AI tools might help them get ahead of blazes.
“Fires are part of our life [now] in the mountains, and we have to be prepared for them,” Vail town manager Russell Forrest tells The Verge.
“Fires are part of our life.”
The collaboration with HPE for wildfire detection came together quickly this year, as Forrest was keeping an eye on a wildfire burning about 30 miles away from his town in July. “One of the things that became very apparent with that fire is that rapid detection of the fire and then the response to that made a big difference and will continue to make a difference in terms of managing future fires where we are,” he says.
Forrest connected with HPE and the company Kamiwaza, which has developed an AI orchestration platform, to ask how AI might be able to help the town. Now, Vail is the first municipality in the US to adopt HPE’s new “smart city solution,“ developed alongside Kamiwaza, Nvidia, and several other tech companies.
The aim is to make it quicker and easier to analyze footage the town already takes using cameras placed on buses and high vantage points on the mountains. Until now, it was primarily people who analyzed those videos trying to spot signs of fire and sometimes being stymied by whether what they saw was smoke or fog. They might have to send people out to a place where there appeared to be a lightning strike to see if it had sparked a blaze.
The new AI-enhanced system can restore higher fidelity in the images if needed and then apply video analytics against it, HPE vice president of AI and hybrid cloud business development Robin Braun explains. It’s been trained to identify lightning strikes and smoke in real time. On the agentic back end, the Kamiwaza platform adds additional context like weather indications (has it been snowing lately or is there a red flag warning in effect?) that might indicate how serious of an event this is and how to respond.
In addition, geospatial data analysis provided by Blackshark.ai is also incorporated into the smart city system. It can utilize drone and satellite images to assess how dry or healthy vegetation is as a measure of fire risk, as well as see how close fire-prone brush is to homes to figure out what needs to be cleared.
... continue reading