IBM Slides despite New AI Dataset for More Accurate Disaster Predictions

In 2024, Bolivia experienced some of the worst wildfires in its history. More precisely, the amount of land scorched was equal in size to Greece. The main causes were land clearing, pasture burning, and a severe drought during what was the hottest year ever recorded worldwide. This disaster is just one of many extreme events featured in a new global dataset called ImpactMesh that has been released by tech giant IBM (IBM) and the European Space Agency (ESA).

Interestingly, ImpactMesh is a massive collection of satellite images that shows the effects of major floods and wildfires from the past 10 years. The dataset includes both “before” and “after” images taken by the Copernicus Sentinel-1 and Sentinel-2 satellites. To make the data more useful, it combines three types of images:

• Regular optical pictures
• Radar scans that can see through clouds and smoke
• Maps showing changes in the land’s elevation

Thanks to this dataset, IBM and ESA improved their TerraMind AI model to better detect disaster zones. By training the model with elevation data and radar images in addition to optical ones, they made it more accurate, even in hard-to-see conditions like heavy smoke or cloud cover. For example, in early tests, the updated model performed at least 5% better than earlier versions. Now that both ImpactMesh and the open-source TerraKit tool are available to the public, researchers and emergency teams can respond more quickly.

Is IBM a Buy, Sell, or Hold?

Turning to Wall Street, analysts have a Moderate Buy consensus rating on IBM stock based on eight Buys, five Holds, and one Sell assigned in the past three months, as indicated by the graphic below. Furthermore, the average IBM price target of $300.58 per share implies that shares are trading near fair value.

See more IBM analyst ratings