Planette Selected by NASA to Develop the First Quantum-Inspired AI System for Extreme Weather Prediction

Estimated reading time: 2 minutes

Extreme weather is on the rise, but even state-of-the-art forecasting technologies struggle to provide accurate, timely predictions for extremes like heavy rainfall, heatwaves, and thunderstorms. To address this limitation, Planette, a leader in long-range weather prediction technology, was selected by the U.S. National Aeronautics and Space Administration (NASA) to develop QubitCast, a next-generation forecasting system, powered by artificial intelligence (AI), that will improve long-range or subseasonal-to-seasonal (S2S) weather predictions (forecasts ranging from two weeks to two years) beyond current limitations, with a focus on extreme weather detection capabilities. The system will provide critical early warning signals for high-impact severe weather, making it possible to plan and prepare for extremes.

Unlike conventional weather modeling, which is limited to 10 days of lead time, Planette’s layering of several physics-based models and AI enables its forecasts to extend beyond this short window. Planette combines atmospheric data with ocean and land inputs and couples this modeling approach to fill a critical information gap between traditional short-term weather predictions and long-term climate projections, up to one year into the future.

With QubitCast, Planette is utilizing algorithms inspired by quantum physics, such as the ability to explore multiple possibilities in parallel. The breakthrough will allow the company to process data more efficiently and uncover hidden weather patterns with greater accuracy, delivering more timely and actionable forecasts without relying on the still-emerging hardware for quantum computing.

"Planette is one of the first companies to take these methods beyond theory and apply them to weather and climate,” said Dr. Kalai Ramea, co-founder and Chief Technology Officer of Planette. “You can think of it like reading the entire history of Earth’s systems all at once. Instead of slowly scanning year by year and missing critical details, our approach allows us to spot anomalies, those needles in the haystack that signal extreme weather events, much faster and more accurately than traditional AI models ever could, while using far less energy."

Current long-term weather forecasts utilize massive physics-based simulation models that are powerful but require enormous computational resources, making them expensive and energy-intensive. Meanwhile, existing AI models that attempt to understand Earth's interconnected systems often become overwhelmed by complex, high-dimensional data, requiring even more computational power and time.

The technology used for QubitCast will require less computing power and use less energy, while distilling complex information into a simpler form, preserving critical details. The quantum-inspired approach enables the system to uncover hidden patterns in how Earth’s systems interact, thereby improving the early detection of extreme weather events and providing actionable warnings.

“Too many critical decisions are made in the dark because reliable long-range forecasts simply haven't been available,” said Dr. Hansi Singh, co-founder and CEO of Planette. “QubitCast changes that equation by making S2S forecasting not just more accurate but practical to deploy at scale when making highly consequential decisions. Whether it's farmers planning crop cycles, emergency managers preparing for hurricane season, or energy companies managing grid stability, this technology provides the lead time needed to act proactively rather than reactively.”

This latest Small Business Innovation Research (SBIR) grant recognition by NASA builds on Planette's growing momentum in advancing weather intelligence technology. Earlier this year, the company was awarded a Phase I SBIR grant from the National Science Foundation to develop NIVA, its foundational AI model for Earth system analysis. The company also recently launched Eddy, making previously inaccessible long-range weather predictions freely available to the public for the first time.