Google used AI to accurately predict floods up to 7 days in advance

Google has announced a groundbreaking use of artificial intelligence to predict riverine floods up to seven days in advance in some cases. Published in the prestigious science journal Nature, the findings mark a significant development in flood forecasting. By leveraging machine learning models trained on various data points such as historical events, river levels, and terrain details, Google was able to generate localized maps and conduct simulations to accurately predict floods.

While the models achieved high accuracy for specific locations, Google aims to expand this technology to a global scale. The company successfully forecasted floods in 80 countries, benefiting a population of 460 million. These forecasts are accessible through Google Search, Google Maps, Android notifications, and the company’s Flood Hub web app. Google’s efforts have notably improved forecasting in underrepresented regions like parts of Africa and Asia.