Weather is one variable that needs to be tracked live as there can be changes every now and then. Today, there are satellites that help smartphones present weather information as accurately as possible. However, like everything else, this too has scope for improvement. Which is why now, two of the biggest organisations in their respective fields, IBM and NASA, are collaborating on developing an AI that can help in the creation of next level weather tracking applications.
IBM studied the market and found that the AI models that are available for weather forecasting and sensing, behave more like AI emulators and not fully trained AI models, which can lead to inaccuracies and a lack of depth. What IBM wants to create is a foundation AI model that will be capable of much more than just weather sensing and predictions. AI emulators rely on the data set that is available, while a foundation AI model is capable of a lot more.
NASA and IBM are presently brainstorming what they want their foundation AI model to be. Compared to the existing models in the market, the collaboration seeks to create a foundation AI model that has expanded accessibility, faster inference times and greater diversity of data. They also want to improve the accuracy of weather forecasting along with predicting meteorological phenomena, inferring high-res information based on low-res data and “identifying conditions conducive to everything from airplane turbulence to wildfires.”
