Helping Utilities Catch Wildfires Before They Spark with CRWN.ai

CRWN.ai helps utilities prevent wildfires sparked by failing grid infrastructure | https://crwn.ai/ These fires are rare, but they're catastrophic. In 2018, a single failure on PG&E's grid sparked California's deadliest wildfire, killed 85 people, and pushed the utility into bankruptcy. The fire made wildfire liability a board-level risk for utilities everywhere. And it keeps getting harder: transmission lines are aging, wildfire seasons are getting more intense, and rising electricity demand means even more infrastructure needs to get built. Most utilities still catch failing lines the same way they did fifty years ago: a lineman driving the corridor once every one to eight years, looking for problems. Brittany Courvoisier-Nicol (  / brittanycourvoisiernicol  , co-founder and Head of Product, and James Playford (  / james-playford-827995bb  , Chief Operating Officer are working to close this gap at CRWN.ai. CRWN.a builds sensors that clip onto power poles and listen, using acoustic and radio-frequency data to catch a failing insulator months or years before it sparks. Instead of one data point every several years, utilities get continuous, real-time awareness of what's actually at risk, and where to send crews first. MORE Subscribe to our weekly newsletter⁠⁠⁠ for Canadian climate tech funding, news, and trends: ⁠⁠⁠⁠⁠https://climatetechcanada.ca/subscribe ⁠⁠⁠⁠⁠Enjoying the show? Like & comment or leave a review on Spotify or Apple Podcasts! Feedback or guest ideas: hello @ ⁠climatetechcanada.ca WE COVER Why CRWN.ai pivoted out of a defence-tech sister company and back into grid resilience How wildfire liability, EV load, and data centre demand are colliding on utility boards' risk radar right now What it actually takes to get a risk-averse utility to put their hardware on live infrastructure Why the funding model that works for defence doesn't exist yet for Canadian climate tech How CRWN.ai incorporates Indigenous knowledge into wildfire-risk modelling, and where it's outperforming Western data What validating sensor readings against a lineman's lived experience on the ground actually looks like