Wfp’s skai model analyzed nearly 500,000 buildings after the myanmar earthquake and over 300,000…
wfp’s skai model analyzed nearly 500,000 buildings after the myanmar earthquake and over 300,000 after the sri lanka cyclone, turning post-disaster damage maps that took days into same-day outputs, and shenzhen’s flood system scanned 243,000 surveillance images in one storm to flag 1,500+ flood risks. satellite ai and camera ai are the first responders now.
the people who rebuild cities will do it from a damage report written by a model within hours of the shaking.
Context
WFP Innovation's SKAI page says that in 2025 SKAI analysed nearly 500,000 buildings after the Myanmar earthquake and more than 300,000 after the Sri Lanka cyclone. SKAI is a machine learning tool for automatic building damage assessment on aerial imagery, set up with Google Research (WFP Innovation; google-research/skai on GitHub).
A Longgang district government page (Jun 24, 2026) describes AI flood control agents in Shenzhen, a 'digital duty officer' and drones that monitor road flooding.
The 500,000 and 300,000 building counts match WFP, and the events were in 2025. The same-day outputs versus days, the 243,000 Shenzhen images in one storm and the 1,500 flood risks were not seen in the sources read, so unsupported here, not refuted. The Shenzhen page read describes the flood control agents in general terms. 'Written by a model within hours of the shaking' is the author's opinion; a World Bank GRADE report on Cyclone Ditwah shows a separate rapid damage estimate for Sri Lanka.
Related work
- WFP Innovation: how WFP uses satellite imagery and machine learning in emergencies ↗SKAI with Google Research.
- World Bank: GRADE report, Cyclone Ditwah 2025, Sri Lanka ↗Dec 17, 2025.
- MDPI: dynamic flood risk assessment in Shenzhen ↗Flood susceptibility framework.
Watch next
- Find the 243,000 images and 1,500 risk points in Shenzhen sources.
Sources
- WFP Innovation: SKAIinnovation.wfp.org
- GitHub: google-research/skaigithub.com
- Longgang government: AI flood control on the front linelg.gov.cn
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 1 October 2026 at 18:35 IST. Sources are the papers and datasets the note draws on.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →