Google DeepMind maps global forests with AI, but plantings remain hard to identify
BROBROK News · Based on reporting by DiarioBitcoin ·
Google DeepMind and the World Resources Institute launched ForTy, a global map that classifies 2020 land cover into six forest-related categories using satellite data and AI. It posts 90.2% overall accuracy, but performs worse at spotting tree crops and planted forests, two categories relevant to deforestation monitoring and supply-chain checks.
Key points
- ForTy uses satellite data and AI to classify 2020 land cover.
- The map covers six forest categories and is built on 10-meter data.
- Overall accuracy reached 90.2%, but errors were common in plantations and tree crops.
- The project could support checks tied to the EU anti-deforestation rule.
- Researchers say regional and classification errors mean the map should be used cautiously.
Why it matters
Maps like this can support traceability efforts in forest- and agriculture-linked supply chains. The weak performance in sensitive categories shows the limits of relying on AI alone for compliance and monitoring.
Original headline: “Google DeepMind mapea los bosques del mundo con IA, pero falla al distinguir plantaciones”
Read the original on DiarioBitcoin
Summary and analysis written by BROBROK News from the original report. Read the full story on the source website. Not investment advice.