AI systems trained around dominant global norms risk overlooking local values and social contexts, creating a need for African developers to shape how the technology understands and serves their communities.
Artificial intelligence is becoming more accessible across Africa, but ensuring that these systems reflect the continent’s cultural values, social norms and everyday realities remains a key challenge for the technology industry.
Businesses, governments and individuals need AI tools that do more than process information; the technology must also understand local languages, cultural practices and social expectations to deliver relevant and useful results.
Jeremy Hitchcock, co-founder and Partner at New North Ventures, believes this is an area that African technology developers and entrepreneurs should pay closer attention to as AI adoption expands.
Speaking at an American Spaces engagement in Accra, under the theme ‘Freedom250: American Excellence in Tech Entrepreneurship and Cyber Security Innovation’, Mr Hitchcock argued that existing AI systems carry cultural assumptions that may not necessarily reflect the societies in which they are being used.
“There is a lot of culturalism that’s baked into how AI functions,” he said, explaining that many AI agents are trained towards similar norms and tend to conform to what he described as a relatively uniform personality.
His observation draws attention to a fundamental concern for Africa’s emerging AI industry, as systems built around the experiences and assumptions of other regions may struggle to serve African users effectively without proper cultural adaptation.
Language translation alone may not be sufficient.
An AI assistant might understand a Ghanaian user’s words but still struggle to interpret the cultural meaning behind a greeting, a local expression, or a particular social interaction. Similarly, advice generated around business practices, family relationships or public institutions may reflect assumptions that are less applicable to African communities.
For businesses deploying AI in customer service, financial services, education and healthcare, these differences could influence how customers interact with technology and how much confidence they place in automated systems.
Mr Hitchcock believes the first step is to identify these differences and understand how they affect the behaviour of AI systems.
“The first thing is being able to come up with how you can tell culturally where your difference is and be able to talk about them,” he explained.
He suggested that developers could examine how AI systems respond to different cultural settings and explore ways of adapting their training to accommodate those differences.
He also pointed to open-source models as a potential avenue for experimentation, allowing developers to modify existing technologies to make them more relevant to particular communities.
Africa has an opportunity to move beyond being a market for imported AI applications and become an active contributor to the technology’s development.
However, cultural adaptation is only one part of a wider challenge involving trust and responsible adoption.
As AI becomes integrated into professional and everyday activities, Mr Hitchcock cautioned against treating its outputs as automatically reliable.
Based on his hands-on work with AI agents, he noted that he often uses the technology to critically assess his own output, pinpoint vulnerabilities, and propose enhancements rather than passively adopting its output.
This highlights the growing need for organizations to keep humans in the loop as they adopt AI tools.
“The AI is supposed to make you more effective and more productive,” he said, arguing that organisations should reconsider how they use the technology if it does not improve productivity.
He also warned against letting AI systems replace independent judgement, comparing them to teenagers with access to extensive information whose work still requires verification.
African businesses must look beyond the speed and cost savings of AI adoption, ensuring that automation does not come at the expense of accuracy, accountability or customer experience.
Developing AI systems that are both efficient and responsive to local cultural and social realities offers Africa a chance to shape the technology around the needs of its people.
