AI weather startup out-forecasts government agencies

A new AI-powered weather forecasting tool released today by startup WindBorne Systems delivers more frequent and accurate predictions on key variables than the world-leading system developed by European governments, thanks to improvements in how sensor data is fed into deep learning models.

Founded by a group of Stanford students in 2019, WindBorne initially focused on building better weather balloons with the goal of selling weather data. However, after the emergence of deep learning models for weather forecasting in 2022, the team realized they could capture more value by developing their own model as well.

WindBorne says the latest version of its model outperforms both the ECMWF’s traditional and AI systems across several variables. According to chief product officer Kai Marshland, one simple way to understand this is that WeatherMesh 6 “is as accurate five days out as a traditional forecast is the day before,” especially for surface temperature measurements.

Today marks the release of the sixth version of that model, WeatherMesh, which the company claims is more accurate than both traditional and AI forecasts produced by the European Centre for Medium-Range Weather Forecasting (ECMWF)—the intergovernmental organization widely regarded by meteorologists as the leading provider of precise weather predictions today.

WeatherMesh 6 generates a forecast every hour, compared to every six hours for traditional models. Its resolution has been reduced to 3 km in Europe and the continental US, where data quality is highest.

Traditional weather forecasts rely on complex physics models that require expensive supercomputers and take significant time to run. AI models—developed by startups and major labs like Google DeepMind—tend to operate faster than physics models, but currently offer lower resolution, fewer variables, and less accuracy over longer time horizons.

Still, AI for weather is improving rapidly and is already being used by major government agencies worldwide. Researchers are working to integrate it into the systems that aggregate weather data and produce public forecasts.

WindBorne benefits from its unique combination of model building and data collection. The company currently has around 400 balloons in flight at any given time, gathering sensor readings from 15 launch sites around the globe. The improvements in its current model come from refinements in how data collected by the balloons is fed into the models.

“I don’t understand, personally, the business model of being an AI-based weather company without a data set advantage,” WindBorne CEO John Dean told TechCrunch.

The ECMWF’s superiority stems from its expertise in “data assimilation”—the process of converting disparate sensor readings into a comprehensive, machine-readable representation of the world. For now, AI weather models depend on datasets produced by the ECMWF and the US National Oceanic and Atmospheric Administration.

But WindBorne and other organizations are working to feed data directly into the models. Joan Creus-Costa, the company’s head of AI, says the direct ingestion of data from their balloons and other sources is the primary reason for the improvement in the new version of WeatherMesh. It took a year of tuning and re-architecting the transformer-based model to deliver these forecasts without losing stability.

“When we started doing data assimilation, we were still very heavily reliant on ECMWF,” Dean said. “I predict today, if we removed ECMWF’s initial conditions, we would actually still do pretty good.”

The company faced a scare last year when a United Airlines jetliner struck one of its balloons. While the plane sustained minor damage, no one was injured, partly because WindBorne complied with US regulations regarding the size of its sensor packages. Now, however, the company has added transponders to its balloons that report their location through the global aviation surveillance system, ADS-B, to reduce the risk of another collision.

WindBorne, which has raised $25 million in venture funding with a reported valuation of $85 million in 2024, sells its balloon data to NOAA (for use in American weather forecasting operations), the U.S. Air Force, and the Navy. The company also sells its forecasts to investors and commodity traders. But Dean says the company remains focused on building out its model and data infrastructure rather than commercial products, partly due to the evolving nature of the information environment.

“I’m not trying to invest a massive team into building a SaaS product, if the way people want consumer information two years from now is through an agent, right?” Dean said.