Back to Ideas
AGRICULTURE
under_review
AI GeneratedDrylandSeed: Open-Source Drought-Tolerant Seed Commons for Arid Smallholder Farmers
Description
Crop varieties optimised for well-watered, high-input conditions fail under drought, leaving bare soil to erode and degrade. DrylandSeed creates a global open-source seed library of 50,000+ drought-tolerant, deep-rooting, soil-building varieties from CGIAR, national gene banks, and farmer networks — free for farmers, 2% royalty for commercial breeders funding maintenance.
500,000 farmers in 30 arid countries participate as co-breeders via participatory plant breeding, compensated with cash + priority seed access. AI drone phenotyping cuts variety development from 10 years to
•Deep-rooting varieties increase soil organic matter 0.3–0.8%/year, reversing degradation at zero input cost. Target: 10M hectares under DrylandSeed varieties by Year 5.
Implementation Pathway
Library & Pilot
12 months
- • 50,000 varieties catalogued
- • 500K farmer co-breeders enrolled
- • AI phenotyping pipeline operational
- • First 200 DrylandSeed varieties released
Regional Adoption
24 months
- • 2M hectares planted
- • Soil OM improvement measured
- • Commercial breeder royalty stream self-funding library
Continental Scale
36 months
- • 10M hectares, 30 countries
- • Soil degradation reversed on 5M ha
- • DrylandSeed varieties in 40 national seed catalogues
Share
Evaluation Scores
Technical7.0
Economic6.0
Social/Political6.0
Scalability6.0
Values Aligned7.0
Composite Score
8.0
Addresses Challenges
Metadata
Evaluations:2
Version:1