Titles, Not Titles: AI for Land and Housing
Walk through the outskirts of Lagos, Kampala, or Nairobi and you’ll see houses rising on every corner — zinc roofs, unpaved roads, plots marked by sticks or tires. Ask who owns the land beneath, and you’ll hear a chorus of different names: a family, a chief, a cousin abroad. The paperwork? If it exists, it’s a crumpled deed in a drawer or a fading page in a council ledger.
The result: disputes, evictions, and billions in dead capital. Hernando de Soto once estimated that Africa sits on trillions of dollars of “dead” assets — land that cannot be used as collateral because ownership is unclear. Without verifiable titles, families can’t unlock loans, developers avoid investment, and cities sprawl chaotically.
This isn’t just a paperwork problem. It’s a trust and data problem. The records are scattered, inconsistent, and often contested.
People should be building:
- AI-Powered Record Digitizers Train OCR models to read smudged paper deeds, handwritten maps, and local land allocation notes. Layer in NLP to interpret different languages and formats.
- Satellite + AI Boundary Mapping Use affordable satellite imagery and computer vision to recognize land use patterns, footpaths, and homestead boundaries. Cross-validate with community records.
- Conflict-Resolution Assistants Build AI systems that can surface similar past cases, relevant laws, and possible compromise frameworks for local mediators.
- Trust-Layer Platforms Combine digitized records with blockchain or other verifiable ledgers. The AI ensures data consistency and flags overlaps before they become disputes.
What success looks like:
- A farmer in Kano digitizes her father’s land papers, verified by AI against satellite boundaries, and finally secures a microloan to expand.
- A community in Nairobi resolves a boundary dispute in days, not years, with AI pulling precedents and showing both sides their overlaps.
- A city council in Accra uses AI-cleaned land data to design roads and water lines, reducing slum proliferation.
Land is wealth. But in Africa, too much of it is invisible to the systems that create value. AI can help us turn disputed plots into trusted titles, and trusted titles into opportunity.