Artificial intelligence is increasingly being explored as a tool for government oversight, but its use in high-stakes areas such as anti-corruption carries risks that African countries may not yet be prepared to manage, according to Dr. Eugene Frimpong, a data analytics and AI specialist.
In an interview with The High Street Journal, Dr. Frimpong stressed that Africa’s underdeveloped legal and regulatory frameworks make it especially vulnerable as governments weigh the adoption of artificial intelligence in public administration. “The rate of innovation always exceeds the pace of governance. Even in Europe and America, laws are struggling to catch up. Africa is even further behind,” he said. Only a handful of African countries, Rwanda, Kenya and South Africa, currently have national artificial intelligence policies in place.
Dr. Frimpong identified transparency as the foremost concern, warning that AI systems often operate as “black boxes” with little clarity on how decisions are reached. “You don’t really know what data is being used or what algorithms are applied. That lack of transparency is concerning,” he said. He added that algorithmic bias remains a risk, since models trained on skewed or incomplete data can unfairly disadvantage marginalized groups such as the poor, the elderly or those with criminal histories.
Security vulnerabilities are another challenge. AI systems can degrade over time without regular audits and backtesting, Dr. Frimpong noted, while data poisoning by malicious actors could manipulate outcomes. “Whatever you put into the machine is what comes out. Garbage in, garbage out,” he said.
While cautioning against large-scale deployment, he argued that Africa should not dismiss AI entirely. Instead, governments could begin with targeted applications, such as procurement oversight, where corruption is most common. Automating smaller-value contracts, he suggested, could improve accountability while leaving complex, higher-value deals to human oversight. “Start small, aim high, but take gradual steps,” he advised.
Dr. Frimpong also urged African policymakers to strengthen privacy protections, safeguard marginalized groups, and establish independent AI governance bodies led by technical experts rather than political appointees.
His comments came in the context of Albania’s recent experiment with an AI system to monitor corruption, a move that underscores both the opportunities and risks of deploying such technology in governance. Albania, ranked 110th out of 180 countries for corruption, hopes the initiative will bolster its bid to join the European Union by 2030.
On a separate note, Dr. Frimpong pointed to Africa’s digital finance boom as a reminder of how technology adoption can reshape economies. Mobile money, he said, has transformed access to payments and small loans. When asked about the possibility of mobile money taking people away from the traditional banking system, he noted that it remains unlikely to replace traditional banks. “It is too early to say mobile money will replace banks. Confidence in the banking system and credit scoring still matter,” he explained.
