Google announced today at an AI-focused press event in New York that it will bring its AI-powered wildfire warning system to the United States, Canada, Mexico, and some parts of Australia. This morning, the firm revealed a number of “AI for good” initiatives, including Google’s attempts to provide flood forecasts to more parts of the world.
The recently revealed method makes use of machine learning algorithms that have been trained on satellite data to detect and forecast the development of flames in real-time. The feature’s initial focus is on assisting first responders in deciding how to put out the fire the most effectively.
Katherine Chou, senior director at Google Research, stated on stage, “Our machine learning models trained using satellite data allow us to spot and track wildfires in real time and anticipate how they will spread.” This enables us to help firefighters and other first responders.”
The function complements a related ML-based flood forecasting function that was first revealed in 2018. The availability of such function is now being increased to 18 more nations.
Google Wildfire Map AI
Google also mentioned that it uses AI algorithms to identify damaged structures during natural catastrophes like hurricanes, as previously documented in an article by Wired. The algorithms give first-phase crisis responders knowledge about the most severely affected locations by evaluating the same satellite images as their wildfire detection technologies.
The technology was initially used by Google in collaboration with the NGO ‘Give Directly’ to track down and distribute donations to low-income households affected by Hurricane Ian. The business claims to have used the models more recently to help government relief operations during the recent flooding in Pakistan. According to Chou, “We intend to continue empowering groups to offer relief to individuals in need faster.”
Also Read: Tesla and Argo AI failures derail driverless technology