WindBorne, which deploys weather balloons to collect data for its AI weather forecasting models, raised a $37M Series B at a $250M post-money valuation
The new deep learning techniques behind LLMs have also given us weather simulations that can run on laptops instead of supercomputers, changing meteorology.
Context & Ripple Effects
WindBorne was already building forecasting around balloon-collected observations, while related coverage flagged that potential NOAA budget cuts could constrain its access to public weather data. Its new financing makes that proprietary-data strategy more consequential.
The company joins a better-funded race with Tomorrow.io, which has raised capital for an AI-driven low-Earth-orbit weather network. The contrast is not just between forecasting models, but between competing ways to gather the data those models use.
First-order effects
- WindBorne gains $37M to support its balloon-based data collection and AI forecasting operation, with a $250M post-money valuation establishing its next financing benchmark.
- Tomorrow.io now faces a newly financed rival pursuing an alternative observation network rather than a satellite-led one.
Second-order effects
- Weather-forecasting rivals relying on public inputs face stronger pressure to differentiate through proprietary observation systems as WindBorne and Tomorrow.io finance their own data-collection layers.
- Potential constraints on public weather-data access, raised in the earlier examination of WindBorne's NOAA exposure, make ownership of observation infrastructure more strategically valuable to forecasting vendors.
Third-order effects
- If this funding pattern persists, AI weather forecasting will increasingly be a vertically integrated business in which durable advantage comes from controlling observations as well as running models.
- Capital allocation may increasingly separate forecast providers with financeable physical data networks from software-only competitors that depend more heavily on shared public inputs.
The trend: AI weather forecasting is becoming a race to pair lower-cost deep-learning models with proprietary systems for collecting atmospheric data.