AI makes weather prediction better. Can WindBorne make it lucrative?

AI's role in nearly every sector continues to grow, and weather prediction is proving to be a critical area where artificial intelligence can make a significant difference. WindBorne Systems, a company at the forefront of this intersection, has announced a substantial $37 million Series B funding round. This investment aims to scale its innovative approach to weather forecasting, which combines a global network of advanced weather balloons with powerful AI models.

The new capital infusion, co-led by Khosla Ventures and Galvanize, with participation from TransLink Capital, Lux Capital, and existing investors, brings WindBorne's total funding to more than $62 million. This financial backing underscores a growing recognition of the urgent need for more accurate and timely weather intelligence, especially as climate change intensifies the frequency and cost of extreme weather events worldwide.

Addressing the Gaps in Global Weather Observation

For decades, weather forecasting has relied on a combination of ground-based stations, satellites, and traditional radiosondes (weather balloons that typically fly for only a few hours). While these methods have improved over time, a significant challenge remains: approximately 85% of the Earth's atmosphere lacks adequate weather observations. These data gaps are most pronounced over vast oceans, remote polar regions, and other inaccessible areas where many critical weather patterns originate. This lack of comprehensive, real-time data severely limits the accuracy and lead time of forecasts, leading to billions of dollars in economic losses annually and, more importantly, a higher risk to human life from unpredicted severe weather.

Traditional numerical weather prediction (NWP) models, while highly sophisticated, often require immense supercomputing resources and hours to generate a single global forecast. This computational intensity can hinder the speed and granularity needed for hyper-local or rapidly evolving weather phenomena. The industry has been actively seeking faster, more reliable approaches that can fill these observational voids and reduce computational overhead.

WindBorne's "Planetary Nervous System" Approach

WindBorne Systems is tackling these challenges head-on by building what it calls a "planetary nervous system." This system is designed to provide continuous, high-precision atmospheric data and transform it into significantly better forecasts. The company's unique technology stack involves two primary components: a global constellation of intelligent weather balloons and a proprietary AI forecasting model.

The Atlas Constellation: Autonomous Weather Balloons

At the heart of WindBorne's data collection strategy is its Atlas constellation, which is currently the world's largest network of autonomous, long-duration weather balloons. These Global Sounding Balloons (GSBs) represent a significant leap from traditional weather balloons. Unlike conventional balloons that burst after a few hours, WindBorne's GSBs can stay aloft for weeks, with some missions demonstrating capabilities for over 75 days.

Each GSB is lightweight, weighing just 1.2 kg, and is equipped with proprietary sensors and an advanced avionics package for real-time communication and navigation. Crucially, these balloons are not passive; they are "intelligent." Their proprietary flight autonomy software allows them to navigate by shifting altitude to different wind layers, enabling them to target specific atmospheric data collection points and capture vertical data slices (atmospheric profiles). This capability allows them to actively gather critical data over oceans, deserts, and other remote areas where observations are scarce, even flying into the centers of storm systems like hurricanes and atmospheric rivers.

The data collected—including temperature, humidity, pressure, and wind data—is transmitted in real-time, often reaching users within an average of seven minutes from collection. This continuous, targeted data collection dramatically improves upon the limited, single-slice data provided by traditional methods.

WeatherMesh: AI-Powered Forecasts

The raw atmospheric data gathered by the Atlas constellation feeds directly into WindBorne's proprietary AI model, WeatherMesh. WeatherMesh (specifically WeatherMesh-6, which was introduced two months prior to the Series B funding announcement) is a transformer-based AI model that runs on GPUs. It is trained on decades of historical atmospheric data and continuously updated by the real-time observations from WindBorne's balloon network.

This combination of cutting-edge AI and unique, real-time data allows WeatherMesh to produce forecasts in seconds, making it up to 100,000 times faster and requiring a fraction of the computational cost compared to traditional NWP models. WindBorne claims that WeatherMesh-6 is currently the world's most accurate weather forecasting model, outperforming both leading AI forecasting systems and traditional physics-based models in publicly available benchmarks. The company also introduced MetaMesh, its first multi-model blend that incorporates WeatherMesh, giving it access to atmospheric signals that other blends cannot replicate.

The $37 Million Series B: What It Means

The $37 million Series B funding round is a significant milestone for WindBorne Systems, bringing their total funding to over $62 million and valuing the company at $250 million. The round was co-led by prominent venture capital firms Khosla Ventures and Galvanize, with additional investment from TransLink Capital, Lux Capital, and previous investors such as Footwork VC, Pear VC, Ubiquity Ventures, Susa Ventures, Jetstream Ventures, and Convective Capital.

This capital will be strategically deployed to accelerate several key initiatives:

• Scaling the Sensing Network: WindBorne plans to significantly expand its Atlas constellation, with a long-term goal of operating thousands of balloons concurrently to achieve complete global coverage.
• Advancing AI Forecasts: The funding will support the continued development and enhancement of the WeatherMesh AI platform, including expanding access to compute resources for training next-generation models.
• Commercial Market Expansion: While WindBorne currently has strong revenue from government contracts with entities like the US National Weather Service, the US Air Force, the US Navy, the National Oceanic and Atmospheric Administration (NOAA), and the National Center for Atmospheric Research (NCAR), this new funding will fuel their push into the commercial weather intelligence market. They plan to build out a dedicated go-to-market team to serve businesses across various sectors.

The company has already demonstrated rapid growth, tripling revenue and the size of its global balloon constellation to over 600 autonomous sensing balloons operating simultaneously in the past twelve months. This momentum positions WindBorne to make a substantial impact on how industries and governments prepare for and respond to weather events.

AI's Growing Influence in Weather Forecasting

WindBorne Systems' success is part of a broader trend where AI is rapidly transforming weather prediction. Companies and research institutions are increasingly leveraging machine learning to process vast amounts of meteorological data, leading to faster, more accurate, and more localized forecasts.

For example, Google DeepMind's GraphCast can generate a 10-day forecast in under a minute on a single computer, outperforming traditional supercomputer models on many metrics. Similarly, the European Centre for Medium-Range Weather Forecasts (ECMWF) introduced AIFS (Artificial Intelligence Forecasting System) in 2024, which became the first fully AI-based system to enter operational service at a major international forecast center. AIFS runs with significantly less energy than its physics-based counterpart while achieving competitive or superior accuracy.

The implications of these advancements are far-reaching:

• Agriculture: Farmers can make better decisions on planting, irrigation, and harvesting, mitigating risks from changing weather patterns.
• Renewable Energy: Improved forecasts for solar irradiance and wind patterns enable better planning and management of renewable energy resources, contributing to grid stability and accelerating decarbonization.
• Logistics and Aviation: More precise weather data allows for optimized routing, reducing fuel consumption and improving safety.
• Disaster Management: Earlier and more accurate predictions of severe weather events like hurricanes, floods, and wildfires provide critical lead time for warnings, evacuations, and resource deployment, potentially saving lives and reducing damage.
• Defense and Government: Specialized meteorological intelligence is vital for mission execution, hardware safety, and national security.

The Future of Weather Intelligence

WindBorne Systems' mission is clear: to eliminate weather uncertainty and help humanity adapt to climate change. By continuously instrumenting the Earth's atmosphere with thousands of autonomous microballoons and refining their AI-powered predictions, they aim to provide unprecedented environmental data and insights.

This latest funding round is a strong indicator of investor confidence in WindBorne's technology and its potential to deliver a more resilient and sustainable future. As extreme weather events continue to pose significant challenges, the ability to forecast with greater speed, accuracy, and detail will become increasingly valuable across all sectors. WindBorne Systems is positioning itself as a key player in this evolving landscape, transforming how we understand and react to the weather around us.

Frequently Asked Questions

How do WindBorne Systems' weather balloons work?

WindBorne Systems uses a constellation of autonomous, long-duration weather balloons, called Global Sounding Balloons (GSBs), which can fly for weeks at a time. These intelligent balloons can actively change their altitude to navigate different wind layers and collect real-time atmospheric data, including temperature, humidity, pressure, and wind, in areas often underserved by traditional weather observations.

What is WeatherMesh, and how does it improve forecasts?

WeatherMesh is WindBorne Systems' proprietary AI forecasting model. It uses a transformer-based AI architecture, trained on decades of historical weather data and continuously updated with real-time information from WindBorne's balloon network. This unique data input and AI processing enable WeatherMesh to generate forecasts significantly faster and more accurately than traditional physics-based models and even other leading AI models.

Who are the main investors in WindBorne Systems' Series B round?

The $37 million Series B funding round for WindBorne Systems was co-led by Khosla Ventures and Galvanize. Other participants included TransLink Capital, Lux Capital, and existing investors.

What impact will WindBorne Systems have on industries?

WindBorne Systems aims to provide more accurate and timely weather intelligence that can benefit a wide range of industries. This includes optimizing operations in agriculture, enhancing grid management for renewable energy, improving logistics and aviation safety, and providing crucial lead time for disaster preparedness and response.