AI That Works for the 90%: Principles for Real Adoption

Over the past few essays, we’ve looked at AI through the lens of adoption for three key groups in food systems:

Ag-input producers, squeezed between distributors and multinationals, Farmers, making daily decisions with little margin for error and Post-harvest companies, struggling to preserve value in a system that loses up to 40% of food after harvest.

In each case, the pattern was clear: technology alone is not enough. Adoption hinges on whether AI passes the test of the Four Pillars—Ease of Use, Implementation Fit, ROI Clarity, and Context Relevance.

But there’s a bigger truth here: most of the world’s food is grown, sold, and moved by small and mid-size actors. They make up 90% of the food system that doesn’t have deep pockets, IT departments, or global reach. If AI is to truly transform agriculture, it must work for them.

Principles for AI in Food Systems

1. Design for Simplicity

   
   

   If a farmer can’t use it on WhatsApp, or a pack-house manager can’t operate it with a smartphone, adoption will stall. Technology must hide complexity, not add it.

   
   

2. Integrate, Don’t Replace

   
   

   AI should fit into existing workflows—sales systems, farm routines, logistics networks—rather than demanding new infrastructure or radical behavior change.

   
   

3. Show the Money (or the Savings)

   
   

   ROI must be obvious. Whether it’s lower spoilage, reduced costs, or higher yields, the benefit must be visible and measurable within a season, not years down the road.

   
   

4. Respect the Local Context

   
   

   Agriculture is not uniform. What works for an Iowa corn farmer may not work for an Indian smallholder or a Mediterranean cooperative. AI must adapt to local crops, climates, and cultures.

   
   

5. Rethink the Business Model

   
   

   Especially for farmers, affordability is the toughest nut to crack. Assume they won’t pay directly. Instead, capture value through input companies, cooperatives, insurers, or governments, who have more capital and stronger incentives.

   
   

6. Build Trust and Transparency

   
   

   Farmers and SMEs are wary of being treated as data sources rather than beneficiaries. Clear communication about data use, privacy, and value sharing is essential for long-term adoption.

   
   

From Hype to Impact

The last decade of ag-tech has been heavy on innovation and light on adoption. That’s why we’ve seen so many startups struggle, and why VC funding has slowed since 2022. The next wave will be different. Success won’t come from dazzling technology—it will come from solving the right problems in the right way for the right people.

If innovators, investors, and policymakers embrace these principles, AI can become a genuine force multiplier for the 90% of the food system that needs it most. And if not, it risks remaining another chapter in the long history of solutions in search of problems.