Artificial intelligence is being used to detect wildfires earlier in the western United States, with cameras spotting smoke and alerting authorities before 911 calls are made. In March, an AI camera in Arizona's Coconino National Forest detected the Diamond Fire, allowing firefighters to contain it at 7 acres.
Utilities and fire agencies are deploying the technology across the region. Arizona Public Service has nearly 40 AI smoke-detection cameras and plans 71 by summer's end. Xcel Energy in Colorado has 126 cameras, aiming to cover seven of eight states it serves by year's end. California's ALERTCalifornia network has about 1,240 AI-enabled cameras.
The systems use high-definition cameras, satellite data, and AI monitoring. Human analysts verify alerts, reducing false positives. ALERTCalifornia's founder said the AI often beats 911 calls. In some cases, firefighters have responded and extinguished fires before any 911 call was received.
Pano AI, a provider, charges roughly $50,000 per camera annually, including fire risk analysis and a 24/7 intelligence centre. The company detected 725 wildfires in the US last year. Customers include forestry operations, government agencies, and utilities.
Challenges include cost and false alarms. The technology does not recommend actions; human decision-making remains essential. In populated areas, people still spot and report fires quickly, but AI is particularly useful in remote regions where fires might go unnoticed for longer.
