Stanford's Tempov Foundation Model Predicts Village Wealth from Satellite Imagery Alone
anselm · x · 2026-07-25
Stanford researchers recently published Tempov, a geospatial foundation model trained on three million pairs of Landsat images spanning two decades. It processes raw satellite imagery to predict asset wealth at the village level across entire continents in near real-time.
Benchmark numbers show the model explains 87% and 74% of the variation in household wealth in Malawi and Mozambique respectively, using just six spectral bands. It operates entirely without census forms, field enumerators, or mobile phone metadata.
Addressing a harder problem than previous static models, Tempov is designed to track actual wealth changes over time rather than just capturing a single static snapshot.
More from Research
- AI slop is already clogging PR review and weakening the credit system behind science — rbhar90 · 2026-07-27
- ICML 2026 oral paper replication scores stay middling after a stricter re-scoring — profjamesevans · 2026-07-27
- Long-running agents will need immutable event logs, this thread argues — sebpaquet · 2026-07-27
- Seed IQ navigates Doom II, prompting questions about benchmarks beyond ARC-AGI — Fit_Transition8824 · 2026-07-27
- Agentic Data Science in Practice: Agents Write Code but Answer Wrong Questions — hugobowne · 2026-07-27
- A concise canon of foundational papers in ML, systems, NLP, speech, and audio — deliprao · 2026-07-27