Urban functional observations are unevenly distributed worldwide, limiting analysis of how places are organised and used. AlphaEarth Foundations (AEF) provides a globally consistent physical representation. Whether this representation can support transfer of functional knowledge to locations with s…

I’ve written before about image similarity with pgvector and the usefulness of embeddings in geospatial workflows. The ability to represent observations as vectors gives us another way to search an…

10-12 March 2027

Google DeepMind and Google Cloud are giving 200 European researchers AlphaFold 3, AlphaGenome, WeatherNext, AlphaEvolve and AlphaEarth, plus $3M in credits. What each does and how to try them.

On October 14, 2026, Google announced that Google DeepMind and Google Cloud are providing up to 200 European scientific researchers with direct access to the company’s frontier models, together with $3 million in compute credits. Under the pledge, researchers will gain direct access to Google’s scientific tools, including AlphaFold 3, WeatherNext, AlphaGenome, AlphaEvolve, and AlphaEarth Foundations. The announcement was authored by Maureen Costello, vice president for the UK, Ireland and Sub-Saharan Africa at Google Cloud. Google said the commitment supports the European Union’s AI for Science work by making some of its frontier models available through the RAISE initiative, which the company described as seeking to help bridge the compute gap European scientists sometimes face. By delivering compute power and tools directly to research teams, Google said, the effort aims to accelerate the pace of European innovation and reinforce the continent’s scientific competitiveness, with researchers equipped to tackle challenges ranging from predicting extreme weather patterns to mapping molecular structures. The European AI in Science Strategy The concept of RAISE was unveiled in the European Strategy for AI in Science, which the Commission launched to reinforce Europe’s technological and scientific leadership and competitiveness by harnessing the potential of AI technologies in...

Accurate and timely crop type mapping offers a wide range of benefits, from supporting annual crop inventories and agricultural monitoring to enabling sustainable resource management and a deeper u...

Fits species distribution models and maps habitat suitability at up to 10 m resolution from occurrence records alone, using the 'AlphaEarth' Foundations satellite embeddings (Brown et al. 2025) &lt;<a href="https://doi.org/10.48550%2FarXiv.2507.22291" target="_top">doi:10.48550/arXiv.2507.22291</a>&gt;. The embeddings, 64 values per pixel per year from a geospatial foundation model, replace environmental layers, so none need to be sourced or aligned. Provides tools to format occurrence records, place pseudo-absences, train and evaluate an ensemble of machine learning models, and export habitat-suitability rasters. Sampling, model training and prediction all run on 'Google Earth Engine', which requires a free account for noncommercial use.