When OpenAI CEO Sam Altman recently told the US Federal Reserve that the world may be on the verge of a “fraud crisis,” his warning might have sounded distant to many in Ghana. Yet for a nation where mobile money, or MoMo, fuels everyday commerce and MTN serves millions of subscribers, Altman’s words could not be more urgent.
Altman described a near-future scenario in which artificial intelligence has “fully defeated most of the ways that people authenticate currently, other than passwords.” He added, “A thing that terrifies me is apparently there are still some financial institutions that will accept a voice print as authentication for you to move a lot of money… you say a challenge phrase, and they just do it.”
While his audience consisted of central bankers and US financial executives, the implications for Ghana, especially given recent lapses in telecommunication security, are stark. In April and May 2025, MTN Ghana confirmed that it was investigating a cybersecurity incident affecting roughly 5,700 customers, as personal data may have been exposed. The company urged calm, assuring the public that MoMo wallets remained safe, while regulators including the Data Protection Commission, Cyber Security Authority, and National Communications Authority launched coordinated investigations.
The Cyber Security Expert Association of Ghana (CSEAG) was swift to condemn MTN’s handling of the incident, arguing that it wasn’t merely a question of numbers or corporate reputation but rather the personal data and trust of thousands of Ghanaians that were at stake. The group also warned of potential insider involvement and called on the company to embrace full transparency, board‑level accountability, and a permanent in‑house cybersecurity team.
Public sentiment echoed suspicions. On Reddit, one user recounted a deeply manipulative scam targeting his mother, in which callers posed as an MTN customer service agent and cited detailed transaction history, including her SIM activation code and old phone number, before convincing her to change her MoMo PIN, resulting in the loss of over GH¢9,500. Many commenters suspected insider access or collusion within MTN itself. Another user observed persistent scam calls claiming to be from MTN, mostly targeting MTN users and often conducted in Twi, raising troubling questions about how fraudsters knew so many personal details.
Against this backdrop, Altman’s warnings take on a deeply local resonance. Fraudsters using AI‑powered impersonation could soon mimic the voices of trusted bank officials, MoMo agents, or even family members, to trick users into authorizing transactions. Where once a suspicious phrase or unfamiliar tone might prompt a callback or hang‑up, deepfake voice or video could erase that hesitation entirely.
This is no longer science fiction. A few years ago, a scammer needed to rely on crude tricks, fake SMS messages, hurried phone calls, or scripted lies that a careful listener could sometimes catch. Now, with off‑the‑shelf AI tools, a scammer in Kumasi or Lagos can feed a few seconds of someone’s voice into an app and generate an eerily accurate copy, ready to deliver whatever lie they need. The same technology can build a video “FaceTime” call that looks convincing enough to fool even a cautious MoMo user.
Experts warn that the barrier to entry for such schemes has collapsed. “It used to take a whole studio of Hollywood effects to fake someone’s identity,” a Ghanaian cybersecurity analyst observed recently. “Now, you need nothing more than a laptop and an internet connection.”
The effect on scams could be devastating. Fraud rings operating in Ghana have already shown a knack for innovation, constantly shifting from fake “MTN promo” messages to phishing links and insider leaks. With AI, they gain a weapon that not only tricks victims more easily, but also overwhelms traditional defenses. Imagine a MoMo subscriber getting a video call from what looks like an MTN agent in uniform, or hearing their pastor’s cloned voice asking for an urgent donation.
AI also allows scams to scale. A fraudster no longer needs to make hundreds of manual calls, one at a time. A single operator could generate thousands of personalized, convincing voice notes in Twi, Ewe, or Ga, each one sounding like it came from someone the victim trusts. Fraud becomes not just smarter, but faster, more targeted, and harder to trace.
For Ghana, a country that has leapfrogged traditional banking with MoMo, that’s a chilling thought. Fraud has already cost the digital payments sector millions of cedis; with AI, the numbers could spiral. Worse, it risks eroding the hard‑won trust that MoMo has built over the past decade, trust that underpins school fees, market payments, and even remittances from abroad.
Ghana’s mobile money industry processed more than GH¢1.9 trillion in transactions last year alone. The service has been hailed as a financial inclusion success story, bringing the unbanked into the formal economy and reshaping how salaries are paid, bills are settled and markets trade. But that trust is fragile. A surge in AI‑enabled fraud could undermine confidence in MoMo, slowing the country’s push toward a cash‑lite economy.
The implications extend well beyond technology. Regulators like the National Communications Authority and the Cyber Security Authority, as well as telcos and banks, may soon have to rethink what “secure” means in a world where a voice on the phone or a face on a video call can’t be taken at face value.
Altman, for his part, says OpenAI isn’t building tools for impersonation, but he admits the race is on to defend against them. He is backing a controversial project called The Orb, which claims it will offer “proof of human” identity online.
“I am very nervous that we have a significant, impending fraud crisis,” he cautioned.
For Ghana, where so much of daily commerce now rests on the convenience, and trust, of a mobile phone transaction, that warning may be less about a distant future and more about a challenge waiting just around the corner.
