There is a comfortable story about why African farmers cannot get loans. It says they are risky. Informal. Hard to reach. Unbankable.
SANDI AI Technologies just won US$50,000 for demonstrating that the story is mostly wrong — and the Ugandan startup’s Grand Prize at the GoGettaz Agripreneur Prize 2026 in Kigali ought to prompt some reflection in the financial sector.
Consider what actually happens when a smallholder farmer applies for credit. She has land in production. She has years of experience reading her soil and her season. She has buyers who want what she grows. She is turned down because she cannot produce an employment record, a title deed, or an asset the lender can seize.
Nothing in that sequence is a judgement about the farmer. It is a judgement about the form she was asked to fill in. The application was designed for a salaried employee in a city and never redesigned for anyone else. Then the resulting rejection gets relabelled as risk.
This is the assumption SANDI AI, founded by Nabakka Sandra, is attacking. Its technology uses artificial intelligence to understand farmers on their own terms — productivity, needs, potential — and to route that understanding into lending decisions that do not depend on collateral. The starting position is that the information required to lend responsibly already exists. It has simply never been collected in a usable form.
“Farmers should not be limited by systems that were not designed around their realities,” Sandra says. It is a polite sentence carrying a sharp point. The exclusion was a design outcome, and design outcomes can be changed.
The cost of not changing them is easy to underestimate. Every farmer who cannot buy inputs produces less than her land allows. Multiply that across millions of holdings and the shortfall stops being individual misfortune and becomes a national output problem — in food supply, in export earnings, in rural employment. Capital sits idle on one side; productive capacity sits underused on the other. Between them is an assessment method nobody updated.
Which is why the interesting part of this award is not the prize. It is the precedent. If a startup from Kampala can build a credible model for lending to farmers without collateral, the claim that it cannot be done stops being a fact about African agriculture and becomes a choice by the institutions that keep declining to try.
SANDI AI will use the money to build further and reach more farmers. The rest of the sector has a different task: to stop describing an unwillingness to measure as an absence of creditworthiness.
