The world is facing an unprecedented climate crisis. With increasingly extreme temperatures, humans will have to deal with droughts, deforestation, more intense weather phenomena, and many other calamities. But what if we told you that artificial intelligence could be the solution?
Engadget has reported on a technological project involving NASA, IBM, and HuggingFace, aimed at creating a ChatGPT-like AI capable of analyzing vast amounts of data in seconds. This project responds to a significant need from NASA, which estimates generating nearly 250,000 terabytes of data only in its scientific missions for 2024.
These three major companies are joining forces to develop an open-source geospatial model that will serve as the foundation for an AI focused solely on climate and natural sciences. The AI will be capable of predicting crop yields, tracking deforestation, and calculating greenhouse gas emissions, among other functions.
The project involves the implementation of IBM’s AI, Watsonx.ai, which will serve as the base model using data collected by the Harmonized Landsat Sentinel-2 satellites over the course of a year. The ESA‘s Sentinel-2 satellites are designed to capture high-resolution images of land and coastal regions in 13 spectral bands.
HuggingFace will host IBM’s model on its open-source AI platform. IBM claims that their team was able to improve the model’s performance by 15% by refining it with “labeled data for flood and fire damage mapping,” and using only half the usual amount of data.
NASA has already announced its development of a ChatGPT-like artificial intelligence for space missions, showcasing the great interest of the U.S. space agency in AI.
Some of the links added in the article are part of affiliate campaigns and may represent benefits for Softonic.
