UNICEF has said that artificial intelligence (AI) will not succeed in rural Nigeria unless underserved communities actively participate in designing digital health tools.
The Chief of Health at UNICEF Nigeria, Dr Maharajan Muthu, said this on Monday in Abuja at the Artificial Intelligence in Health Summit 2026.
Muthu said predictive analytics, geographical mapping and existing health data should be combined to identify children and poor households left behind by health systems.
According to him, technology designed solely in urban centres without rural community engagement cannot adequately address barriers to frontline healthcare delivery in remote areas.
“AI will fail horrendously at reaching anybody that is rural, remote, disadvantaged as long as those people are not in the room when you conceptualise your system.
“We must engage directly with communities before designing AI solutions, rather than deploying urban-developed systems into rural environments without considering local conditions.
“Combining geographical information systems with health data helps health authorities identify remote populations facing severe information gaps and limited service delivery,” he said.
Dr Olasupo Oyedepo, Managing Partner at Oyedepo & Associates, said technological applications required effective governance, local expertise and functional institutional processes to succeed.
Oyedepo said offline clinical decision-support tools enable health workers in areas with limited connectivity to assess symptoms, analyse disease patterns and accurately predict local healthcare needs.
He said AI models should be trained using indigenous Nigerian languages and cultural contexts to serve underrepresented rural populations effectively.
“AI, per se, will not deliver healthcare services, as we fundamentally need human health workers to provide clinical care.
“Offline applications can support frontline workers in rural communities where internet connectivity is unreliable, improving service delivery when combined with adequate personnel.
“Deploying AI systems developed primarily in English risks excluding communities whose native languages are poorly represented in global training datasets,” he said.
Dr Kelechi Okoro, founder of Healthertainer, said integrating AI into healthcare was intended to augment, rather than replace, the vital work of frontline health personnel.
Okoro said predictive algorithms help health authorities allocate financial resources efficiently, identify vulnerable populations and plan targeted public health interventions.
She emphasised that clinical decision-support systems provide real-time guidance to healthcare providers working in underserved primary healthcare centres across the country.
“Artificial intelligence can analyse disease patterns, predict health needs and help authorities determine the exact interventions and resources required.
“Clinical decision-support tools, including offline applications, help health workers assess symptoms and make informed decisions in rural communities.
“AI cannot independently deliver healthcare, meaning human health workers, functional facilities and adequate infrastructure remain essential,” she said.
