you may have Lately, I’ve noticed a decline in AI in weather apps. As companies race to infuse artificial intelligence into all their products, even the humble weather app is making its way.
The Weather Company, which operates the Weather Channel, today released an improved version of its Storm Radar app with AI-powered Weather Assistant. This allows users to customize how forecasts and weather maps are displayed by switching between layers such as radar, temperature, and weather conditions such as wind and lightning.
You can also sync with other apps like Calendar to send text notifications and weather summaries to tie information about upcoming weather into your daily plans. If you’re interested, you can also paste audio and speak like an old-time radio weatherman. Like most weather apps, it gets its data from the National Oceanic and Atmospheric Administration (NOAA) and the National Weather Service (NWS).
The app costs $4 per month. It’s currently only available on iOS, but the company says an Android version is coming eventually.
“We wanted to build an experience that would level up weather for everyone, from the casual observer to the seasoned storm chaser,” says Joe Koval, senior meteorologist at the Weather Company. “If you’re looking for advice on when the weather will be good for walking your dog tomorrow, you no longer have to sift through tons of different weather data elements to try to figure out the answer to that question yourself.”
Of course, the weather forecast can already be found on your mobile phone. Android and iOS devices typically display weather prominently next to the time. Both Google and Apple are integrating weather apps directly into smartphones. Since then, AI capabilities have been incorporated to provide insights and overviews of the upcoming day.
But there are plenty of third-party weather apps, including Storm Radar, Carrot Weather, Rain Viewer, and Acme Weather, an app from the former Dark Sky app creator. New weather apps like Rainbow Weather aim to be AI-first. Weather services are also directly integrated into AI chatbots, like Accuweather, which recently launched apps directly on OpenAI’s ChatGPT.
“Everyone has their own ideas about what they want from a weather app, what kind of data they’re interested in, and how that data should be presented,” said Adam Grossman, founder of the DarkSky app. “How do you build a single weather app that works for everyone?”
DarkSky, one of the most popular iOS weather apps, was acquired by Apple in 2020 and integrated into the Apple Weather service. Grossman eventually left Apple and started Acme Weather with the goal of creating a weather forecasting service that better communicates forecast uncertainty.
“No matter how good your predictions are, they can still come off,” Grossman says. “This is something that weather apps have traditionally not done well. Our approach is trying to find a way to add back those pieces of context.”
Repositories of weather information typically come from government sources such as NOAA and other global weather services, which collect data from weather satellites, radar, weather balloons, and ground-based instruments. All this data is input into a weather prediction model that simulates the physics of the atmosphere. These predictions are often generated by resource-intensive supercomputers, but machine learning models have cut down on that processing and made predictions faster. (However, it may also have lower accuracy, which can be explained by comparing multiple models.)
Weather apps like Storm Radar and Acme Weather transform that rich information by backing and compiling models to help create high-resolution maps and visual representations of the data, and this is where AI is especially useful.
