- DBG’s ‘Boa Mi’ AI Platform Targets One of Ghana Agriculture’s Hardest Financing Problems
Development Bank Ghana is betting on artificial intelligence to tackle one of the most persistent but less visible barriers to agricultural finance: the lack of timely and reliable information connecting farmers, agricultural advisers and lenders.
At its ICT Sector Financing Roundtable in Accra, DBG Deputy Chief Executive Officer for Investments Michael Mensah-Baah unveiled work on an AI-powered agricultural advisory platform intended to give smallholder farmers, extension officers and financial institutions easier access to practical, locally relevant farming information.
The initiative, being developed with Opportunity International and the Ghana Incentive-Based Risk Sharing System for Agricultural Lending, or GIRSAL, represents an attempt to use technology not simply to digitise agriculture but to address some of the information asymmetries that make the sector difficult to finance.
The project emerges from roughly 18 months of work by DBG examining how technology can be converted from an innovation concept into a tool capable of addressing practical economic constraints.
“Agriculture remains at the heart of the Ghanaian economy, and more importantly, at the heart of the livelihoods of millions of Ghanaians,” Mr Mensah-Baah said. “But for many smallholder farmers in Ghana, one of the biggest challenges is not the willingness to farm.
It is access to the right information at the right time in a form that is practical and easy to use.”
The platform, identified in the document as “Boa Mi,” meaning “help me” in Akan, is being designed around WhatsApp rather than requiring farmers to download a separate application or learn an unfamiliar digital interface.
That design choice could prove more important than the sophistication of the underlying artificial intelligence because adoption ultimately depends on whether farmers can use the technology easily in their normal routines.
“You don’t need to download an application. You don’t need to learn a new technology. You don’t need to navigate a complicated digital platform. You simply need to know how to use WhatsApp,” Mr Mensah-Baah said.
Through WhatsApp, users can ask agricultural questions and receive answers intended to reflect Ghanaian farming conditions, with commodity-specific information supported by GIRSAL’s agronomic expertise.
The approach builds on an earlier DBG and Opportunity International pilot that used generative AI to support rice farmers and extension officers with agricultural advice.
The central proposition is that artificial intelligence may be useful not because it replaces agricultural expertise, but because it can make that expertise available more quickly and at greater scale.
But DBG’s ambition extends beyond extension services, and this is where the initiative becomes economically more interesting.
Agriculture remains difficult for financial institutions to finance because farmers often have limited conventional collateral, revenues are seasonal and lenders frequently possess incomplete information about conditions on individual farms after loans have been disbursed.
The resulting information asymmetry means a farmer may understand production risks intimately while the lender financing that farmer has comparatively limited visibility into what is happening on the ground.
Mr Mensah-Baah argues that the platform could help narrow that gap by providing another layer of information around production conditions and farming practices.
“A loan officer is trying to make a responsible decision about whether to lend to that farmer or not, even though he has not been to the farm,” he said. “This platform can help provide that additional layer of information.”
The implication is potentially significant because Ghana’s agricultural credit problem cannot be solved simply by instructing banks to lend more.
Financial institutions need credible mechanisms for assessing risk, monitoring borrowers and identifying emerging production problems before those problems become repayment failures.
“We are not simply creating an agriculture advisory tool. We are creating a bridge between agriculture knowledge on one hand and agriculture finance on the other hand,” Mr Mensah-Baah said.
That bridge is where DBG’s development-finance mandate could become particularly relevant. Rather than replacing commercial banks, a development finance institution can help create the risk-sharing architecture, information systems and partnerships needed to make sectors traditionally regarded as difficult to lend to more attractive to private capital.
GIRSAL already provides risk-management support for agricultural lending, while DBG says it has channelled more than GH¢288m in agricultural-sector loans through participating financial institutions.
The AI initiative adds information to that financing architecture, and early pilot numbers suggest farmers are willing to use the service. During a rice-farming pilot, the platform generated more than 8,500 advisory queries in three months, with reported user satisfaction above 75.00%, according to Mr Mensah-Baah.
“For me, the most important statistic is not the number of queries. It’s what those queries represent,” he said.
“They represent farmers asking for help. They represent a farmer trying to protect his crop. They represent a farmer trying to increase productivity,” he added, saying the questions ultimately reflected farmers seeking to protect investments and household livelihoods.
The next test, however, will be whether engagement with the platform translates into measurable improvements in yields, farm resilience, credit performance or lender confidence rather than remaining primarily a measure of technology adoption.
That distinction matters because AI-generated agricultural advice carries its own risks. Recommendations must remain accurate, locally relevant and sufficiently dependable for farmers making decisions involving seeds, crops, inputs and incomes, while complex agronomic problems may still require physical farm inspections and human extension officers. The platform should therefore be judged as a complement to agricultural extension and credit assessment rather than as a substitute for professional judgement.
The partnership model is intended to combine different forms of expertise: Opportunity International brings experience working with farmers and communities, GIRSAL contributes agricultural and risk-management knowledge, and DBG provides development-finance capacity. “Three institutions, different strengths, one shared ambition,” Mr Mensah-Baah said.
For DBG, the larger experiment is whether technology can become part of the infrastructure of financing itself rather than another standalone digital pilot that generates interest without materially changing capital allocation.
“As we prepare for the official launch, I believe we should look at this not as the launch of another technology product, but as the beginning of a new way of supporting the growth of the agricultural sector in Ghana,” Mr Mensah-Baah said.
That is an important benchmark because Ghana does not necessarily need more applications; it needs technologies capable of reducing the structural costs, uncertainty and information gaps that keep productive farmers outside formal credit markets. The source document refers to the platform as “Boa Mi” earlier and later as “Buame”, so the precise official branding should be confirmed ahead of publication.
The bigger question is whether better information can genuinely make agriculture more bankable. If the platform eventually helps lenders distinguish viable farmers from higher-risk borrowers, identify production problems earlier and improve loan performance, AI could become part of the financial infrastructure connecting Ghanaian agriculture to capital rather than merely an advisory convenience.
As Mr Mensah-Baah put it, when farmers ask for help, DBG wants the answer to be that it “is here to help” but the enduring measure of success will be whether that help ultimately produces more productive farms, stronger loan portfolios and greater flows of sustainable agricultural finance.
