background image

Google launched an AI model that functions like a virtual satellite. Here’s how it works

author image

By illuminem briefings

· 3 min read


illuminem summarises for you the essential news of the day. Read the full piece on Euronews or enjoy below:

🗞️ Driving the news: Google DeepMind has unveiled AlphaEarth Foundations, an AI model capable of producing highly detailed, near real-time maps of Earth’s land and coastal waters
The system integrates trillions of images from diverse sources — including satellites, radar, laser mapping, and climate simulations — to deliver ecosystem data with a resolution of up to 10 m²
Initial tests from 2017–2024 show it outperformed similar models in land use classification and surface property estimation, reducing error rates by 24%

🔭 The context: For decades, Earth observation has relied on multiple satellites operated by agencies such as NASA and the European Space Agency (ESA)
While these provide critical data for tracking environmental change, combining heterogeneous datasets has been a technical challenge
Recent advances in AI have made it possible to fuse and interpret large-scale geospatial data more efficiently, enabling applications in deforestation monitoring, water management, and urban planning

🌍 Why it matters for the planet: AlphaEarth Foundations could significantly accelerate environmental research and policymaking by offering high-resolution, frequently updated global datasets
Such capabilities can support precision agriculture, disaster preparedness, climate impact assessment, and biodiversity conservation
However, equitable access, data transparency, and safeguards against misuse will be essential for its responsible application

⏭️ What's next: Google (see sustainability performance) is making AlphaEarth Foundations data available via Google Earth Engine, aiming to expand adoption among scientists, NGOs, and policymakers
Over 50 organisations — including Brazil’s MapBiomas — are already using it for ecosystem monitoring
Broader uptake could influence national climate adaptation strategies and international environmental reporting frameworks within the next 2–3 years, especially as AI-driven geospatial tools gain traction in global sustainability initiatives.

💬 One quote: “This gives us new options to make maps that are more accurate, precise and fast to produce, something we would have never been able to do before,” — Tasso Azevedo, founder of MapBiomas

📈 One stat: AlphaEarth Foundations can generate environmental data at a spatial resolution of 10 m², requiring less storage than comparable AI systems while improving accuracy by 24% over existing models

See on illuminem's Data Hub™ the sustainability performance of Google and its peers Microsoft, IBM, and Amazon

Click for more news covering the latest on green tech and corporate governance

Did you enjoy this illuminem voice? Support us by sharing this article!
author photo

About the author

illuminem's editorial team, providing you with concise summaries of the most important sustainability news of the day. Follow us on Linkedin, Twitter​ & Instagram

Other illuminem Voices


Related Posts


You cannot miss it!

Weekly. Free. Your Top 10 Sustainability & Energy Posts.

You can unsubscribe at any time (read our privacy policy)