GoKawiil
Tech News
Swiss weather tech company Meteomatics says its drones can gather atmospheric data more quickly and accurately than traditional weather balloons. Unlike balloons, which launch at fixed times and are discarded after a single use, the company's drones can be deployed on demand to track developing weather conditions like storms.
Mistral has released Timestral, its first family of time-series foundation models, with sizes ranging from 5M to 209M parameters. The company says its largest 209M-parameter model tops all three major public benchmarks—GIFT-Eval, TIME, and fev-bench—surpassing previous state-of-the-art models that used more parameters. The models produce probabilistic, multivariate forecasts and support both past and known-future covariates, with fine-tuning available through Mistral Forge.
Microsoft researchers published a study in Nature describing a machine learning system that corrects bias in existing physics-based and AI weather models, improving forecast accuracy two to six weeks ahead. The tool, now available as open-source software, can run on standard CPUs rather than requiring supercomputers or powerful GPUs.
Researchers describe a probabilistic forecasting method for predicting weather two to six weeks ahead, built on ERA5 reanalysis data from 1979-2026 and evaluated using ECMWF's AI Weather Quest competition framework. The approach forecasts temperature, precipitation and sea-level pressure as probability distributions across climatological quantile bins, scored using the Ranked Probability Skill Score.
The U.S. Air Force has awarded WindBorne Systems a contract worth up to $49 million for atmospheric data collected by its fleet of roughly 700 autonomous weather balloons. The deal, awarded without competitive bidding, lets the Air Force order data by region and time as needed, with $1.7 million in orders placed so far.
Forecasters are tracking three separate tropical cyclones expected to make landfall along North American coastlines within a two-day span. The systems are projected to affect areas of the United States and Mexico nearly simultaneously, prompting heightened monitoring by weather agencies.
A low pressure system in the western Gulf of Mexico is forecast to strengthen into a tropical storm, with forecast models giving it a decent chance of reaching hurricane strength. No hurricane has formed in the Atlantic yet this season, which would be unprecedented in the modern record dating to World War II, when aircraft and satellite monitoring began.
A CNET reviewer spent several months testing Amazon's Alexa Plus, Google's Gemini for Home and Apple's Siri AI for weather forecasting accuracy in a high-desert city with microclimates. The test found Alexa Plus gives the longest, most detailed responses—including wind, air quality and 'feels like' temperatures—but occasionally mispredicted rain and tended to overestimate high temperatures.
Okanagan Specialty Fruits, which farms over 1,250 acres of apple orchards in Washington State, is testing camera-based AI from Canadian firm Vivid Machines to forecast crop yields and optimal harvest dates. The cameras, mounted on tractors, capture images of trees so AI can identify buds, flowers and fruit, producing crop estimates alongside weather-informed harvest timing. Company representative Joel Carter says the system catches small flower buds humans might miss, though forecast accuracy depends on the quality of historical data fed into it.
Prior Software Inc. has released Prior, a predictive intelligence tool that lets users ask questions about future events—from sports and financial markets to personal choices—and receive a forecast report. The platform gathers context through follow-up questions, scans current web information, and reviews a user's past predictions before generating a calibrated probability, weighted scenarios, key drivers, and suggested actions. Pricing ranges from a free weekly prediction to tiered subscriptions up to $499 monthly for unlimited forecasts.
Google has introduced WeatherNext 3, a new AI-driven forecasting system meant to replace or augment traditional numerical weather prediction methods. The system aims to deliver faster, less computationally expensive forecasts while maintaining or improving accuracy compared to physics-based models.
Scientists studying hurricanes have identified a pattern in ocean surface temperature gradients that appears tied to episodes of rapid storm intensification, a phenomenon that has long confounded forecasters. By tracking how sharply temperatures shift across ocean regions a storm passes over, researchers believe they can better anticipate when a hurricane is about to strengthen abruptly.