Regional: Pg&E Uses Artificial Intelligence To Weigh Elevated Fire Risk
Pacific Gas & Electric is increasingly turning to artificial intelligence to refine the accuracy of weather models as it forecasts an above average risk in the wildfire season.
The utility has issued Public Safety Power Shutoffs since 2018 to cut power in portions of its service area, often affecting thousands of customers, to prevent its equipment from sparking wildfires during severe, dry wind conditions.
Article continues below this ad
An AI prediction model uses historical data and machine learning algorithms to increase the accuracy of the company's predictions, PG&E Chief Meteorologist Scott Strenfel said Friday.
The company is preparing for "above normal" fire potential into September, Strenfel said, citing the outlook from the National Interagency Fire Center, which coordinates wildland fire data.
The NIFC study predicts increased flammability in Northern California due to an early growing season that produced abundant vegetation in the region, poor snowpack that will create dry fire fuels and a combination of heat, wind and lightning events.
This fire potential is expected to build on a fire season that has already been above average in California, with Cal Fire reporting 2,854 wildfires that have consumed almost 80,000 acres so far this year.
Strenfel said the number of fires is lower than this time last year, but they have burned roughly 20,000 acres more than the five-year average.
Artificial intelligence is allowing the company to improve the accuracy of its predictive weather models, he said, increasing PG&E's confidence in its ability to prevent fires.
The Bay Area's best free newsletter.
By signing up, you agree to our Terms Of Use and acknowledge that your information will be used as described in our Privacy Policy.
Every time the company rolls out a new model, he said, "we start right back at the beginning and think about what new tools, what new technology can we bring to eke out even more performance?"
"And every time we get better, get more accurate at forecasting the weather, well that makes the PSPS events more precise and just a better experience for everyone."
Copyright © 2026 Bay City News, Inc. All rights reserved. Republication, rebroadcast or redistribution without the express written consent of Bay City News, Inc. is prohibited. Bay City News is a 24/7 news service covering the greater Bay Area.
Related Stories
AI News
Some assembly required, with machine learning
43 minutes ago
AI News
How Artificial Intelligence is Reshaping Global Manufacturing and Value Chains
43 minutes ago
AI News
Hebbia helped popularize AI for Wall Street. Now it's trying to stay ahead of the companies chasing the massive opportunity. The startup, one of the earliest companies to use AI to analyze large collections of financial documents, is overhauling its...
1 hour ago
AI News
Nearly 700 AI agents coordinated Hugging Face attack, says report
1 hour ago
AI News
Marvell Technology Sees AI Connectivity as Next Big Infrastructure Bottleneck
1 hour ago
AI News
Jobs Artificial Intelligence Will Not Be Able to Replace for Now
1 hour ago
AI News
Z.ai shares surge 8% after releasing new AI model running only on Chinese chips
1 hour ago
AI News
The Regulatory Vacuum in AI
1 hour ago