Researchers have developed a new artificial intelligence system that can reliably forecast weather up to six weeks in advance, overcoming a long-standing barrier in meteorology. This AI model, created by Judah Cohen and Lester Mackey at MIT, utilizes a postprocessing technique to learn from its prediction errors by comparing them against historical data. Submitted to the European Center for Medium-range Weather Forecasts, the model reportedly doubled the accuracy of existing forecasting methods for the crucial two-to-six-week subseasonal window. This advancement holds significant potential for sectors like agriculture and energy, and could even improve predictions of extreme weather events, potentially saving lives. AI
IMPACT This AI model's breakthrough in subseasonal forecasting could significantly improve predictions for agriculture, energy, and extreme weather events.
RANK_REASON The cluster describes a new AI model and its performance on a specific task (weather forecasting), published in a scientific journal. [lever_c_demoted from research: ic=1 ai=1.0]
Read on Mastodon — mastodon.social →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
