Banking consultant Dr. Richmond Atuahene wants Ghana’s banks and telecom operators to verify customer identity in real time and let artificial intelligence (AI) screen every mobile money transaction for fraud.
The call comes as fresh data show how far ahead of the country’s defenses digital fraud has moved. Mobile money fraud accounted for 31.5 percent of all recorded fraud cases in the Ghana Bankers Association’s report for the first quarter of 2026, the highest share of any category, while a 2024 study by the E-Crime Bureau found that 65 percent of mobile money users had either been targeted by fraud or fallen victim to it. Recovery offered little comfort: attempted mobile money fraud losses in that quarter reached GH₵143,108, with net losses of GH₵142,038, meaning banks clawed back almost nothing.
For the millions of Ghanaians who now keep more money in a mobile wallet than in a bank account, Atuahene’s plan amounts to a test of whether the system they depend on can keep pace with the people trying to drain it.
His sharpest recommendation targets SIM swap fraud, which Atuahene has separately called one of the most destructive fraud methods hitting Ghanaian mobile money users. He wants banks and mobile network operators to link their systems directly, so a large transaction automatically triggers a check on whether the customer’s SIM card has been replaced in the past 24 to 48 hours. A recent swap would delay the payment or force extra verification rather than letting it through.
He also wants banks to tighten identity checks before they ever open a wallet. Banks and telecom firms should validate every mobile money and digital account in real time against the National Identification Authority’s Ghana Card database, he argues, then add biometric checks that match a customer’s live photograph to their identification record. The goal is to stop fraudsters from registering accounts using stolen identities, forged documents or images generated by AI in the first place.
Atuahene’s broader vision rests on AI handling that screening at scale. Under his model, software would assess every transaction within milliseconds, weighing a customer’s normal spending habits against signals such as a sudden change of device or location, a recent SIM swap, or a payment far larger than the account’s history. Systems could then block a transaction automatically or hold it for extra verification before any money moves, rather than investigating losses after they happen.
Mobile money agents, who handle most cash transactions for Ghana’s wallet users, get separate attention in the plan. Atuahene recommends anomaly detection software that watches agent networks for unusual withdrawal patterns, repeated transaction reversals, abnormal cash withdrawal volumes, and signs that groups of agents are working together to defraud customers.
Atuahene wants centralized systems that track every fraud alert from the first report through investigation, evidence collection and reporting to the Bank of Ghana, arguing that scattered, manual case files make it harder to hold staff and agents accountable.
None of it works in isolation, Atuahene argues, because fraud rings now move across banks, fintech apps and telecom networks within the same scheme. He wants shared intelligence platforms that let banks, mobile money operators, telecom companies and regulators trade fraud data in real time, rather than each institution fighting the same criminals separately. The Bank of Ghana already updated its rules for electronic money issuers in January 2024 to require two factor authentication on transactions and compulsory cybersecurity audits, a step Atuahene’s plan treats as a floor rather than a finish line.
