How scammers are using AI to steal money from Ghanaians: The new tricks everyone needs to know
Artificial intelligence is giving online criminals new ways to deceive people, impersonate trusted individuals and steal money. From cloned voices and fake video calls to convincing investment advertisements and fraudulent messages, scammers are exploiting technology to make their schemes harder to detect. As Ghana's digital economy expands, the question is no longer simply whether an online message is genuine, but whether the person behind it is who they claim to be.
Imagine receiving a telephone call from your brother, mother, employer or close friend asking you to send money urgently.
The voice sounds familiar. The person knows your name, speaks in the same manner and appears to understand a situation that only someone close to you would know.
You do not hesitate. You reach for your phone, open your mobile money application and transfer the requested amount.
Minutes later, you discover that the person who called was not your relative, friend or employer.
It was a scammer using artificial intelligence (AI) to imitate someone's voice.
This is one of the emerging dangers of digital fraud: criminals no longer need to meet their victims, physically handle their phones or even speak to them in their natural voices to make fraudulent requests appear believable.
Artificial intelligence is changing the way online scams are organised. Tools that can generate realistic voices, manipulate photographs, create convincing videos and produce persuasive messages are increasingly being exploited by criminals.
For Ghana, where mobile money, social media, digital banking and online commerce have become important parts of everyday life, the implications are significant.
A fraudulent message that once contained obvious spelling mistakes or sounded suspicious can now appear professionally written. A fake investment advertisement can feature a familiar public figure. A voice message can sound like someone a victim knows.
The danger is that AI is making it easier for scammers to manufacture trust – and trust is often what they need before asking for money.
The scale of the problem: Ghana's growing exposure to digital fraud
The threat is not merely theoretical.
Figures disclosed by Ghana's Cyber Security Authority (CSA) in September 2026 paint a concerning picture of the country's cybersecurity environment.
The Authority recorded 3,876 cybersecurity incidents between January and July 2026, of which 1,818, representing approximately 47 per cent, were linked to online fraud.
The figures came as Ghana continued expanding its use of mobile money, digital banking, electronic payments and online commerce.
The Bank of Ghana's 2025 Fraud Report, cited by the CSA, also showed that reported fraud cases across banks, specialised deposit-taking institutions and payment service providers increased from 16,733 in 2024 to 24,778 in 2025 – a rise of 48 per cent.
The total value at risk was reported at GH¢101 million. Electronic fraud incidents in the payment service provider sector increased by 54 per cent.
These figures cover cybercrime and fraud more broadly. They should not be interpreted as the number of crimes committed using artificial intelligence alone.
However, they provide important context for understanding why the growing use of AI by criminals is a concern for Ghana's digital financial system.
In October 2026, the CSA's Public Relations Officer, Benjamin Madro Avornyotse, told JoyNews that cybercriminals were increasingly adopting AI tools in their operations.
He said investigations had revealed cases in which suspects were using AI to facilitate their activities, creating additional challenges for law enforcement agencies.
The warning is significant because AI is not necessarily replacing traditional fraud methods. It is making some of those methods more convincing, easier to scale and potentially more difficult to identify.
Six ways scammers are using AI to target Ghanaians
1. Cloning the voices of relatives, friends and public figures
Estafas de clonación de voz con inteligencia artificial: ¿Cómo protege
One of the most concerning applications of AI in financial fraud is voice cloning.
Voice-cloning technology can generate speech that resembles the voice of another person using audio samples. Criminals can exploit publicly available recordings, including interviews, social media videos, speeches and voice messages.
The objective is to create the impression that a trusted person is making a request.
Imagine receiving a voice note that sounds like your sister asking for money to settle an emergency hospital bill. Or a call that appears to come from your employer requesting an urgent transfer to a supplier.
The victim may recognise the voice and assume the request is genuine.
In reality, the criminal may be using artificially generated speech to impersonate that individual.
This form of deception is particularly dangerous because many people rely on familiarity with someone's voice as proof of identity.
But a familiar voice is no longer sufficient evidence that the person communicating is genuine.
What Ghanaians should do: If someone calls or sends a voice note requesting money urgently, disconnect and call the person using a number already saved in your contacts. Ask a question that only the genuine person would know, or establish a family verification phrase for emergencies.
Never transfer money solely because a voice sounds familiar.
2. Creating fake videos of politicians, celebrities and trusted personalities
When AI Is Used to Scam Millions: What the ‘Princess Leonor’ TikTok Fraud Reveals About the Future of Financial Crime and How Victims Can Fight Back
Another emerging threat involves deepfake videos.
A deepfake is digitally manipulated or artificially generated audio, video or imagery designed to make someone appear to say or do something they never actually said or did.
In Ghana, the Cyber Security Authority has specifically warned about AI-generated videos featuring prominent personalities, including the President, ministers, Members of Parliament, media houses and media personalities.
The videos have allegedly been used to promote fraudulent investment schemes, financial platforms and unapproved medical products.
A video may show a prominent public figure announcing an investment opportunity and encouraging people to deposit money into a particular platform.
The presentation may look professional. The person's face and voice may appear authentic, and the advertisement may carry the branding of a familiar media organisation.
But the endorsement may be entirely fabricated.
The criminal's objective is to exploit the public's confidence in a recognisable personality to make an illegitimate investment appear credible.
In May 2026, the Ghana News Agency reported that 11 people had been arrested over allegations that they used AI-generated deepfake videos to impersonate President John Dramani Mahama and solicit money and sensitive personal information.
Police said preliminary investigations indicated that the suspects were connected to a wider network involved in fraudulent online content.
The arrests demonstrate that deepfake-enabled impersonation is not simply a hypothetical threat in Ghana.
What Ghanaians should do:
Never invest because a politician, celebrity, pastor or media personality appears to endorse a product in a video. Check the person's verified
communication channels and confirm whether the investment provider is licensed by the Securities and Exchange Commission or the relevant regulator.
3. Producing convincing phishing messages and fraudulent emails
Person checking email security, finding possible phishing attempt
For years, one way people identified online scams was through poor grammar, spelling errors, unusual sentence construction and suspicious email addresses.
Artificial intelligence is weakening that line of defence.
Scammers can use AI writing tools to produce professional-looking messages that imitate the communication styles of banks, telecommunications companies, delivery services, government agencies and online retailers.
A victim might receive a message claiming that their mobile money account has been suspended, their bank account requires verification or a parcel cannot be delivered until a fee is paid.
The message may contain a link directing the recipient to a fraudulent website designed to collect passwords, banking details or mobile money credentials.
In September 2026, the Ghana Cybersecurity Industry Forum warned that criminals were using AI to make phishing messages more convincing by correcting the poor grammar that previously helped people identify suspicious communications.
Its interim president, C.K. Bruce, said the messages could now be well-written but remain fraudulent.
The important lesson is that good grammar is not proof of authenticity.
What Ghanaians should do: Avoid clicking links in unexpected messages.
Instead, visit the official website or contact the institution through its published telephone number. Never provide your PIN, password or one-time password through a link received in an unsolicited message.
4. Using AI to create fake investment opportunities
How Investment Scams and Clone Websites Thrive on Social Media Ads
Investment fraud is another area where AI can make deception more persuasive.
A scammer can create a website, generate promotional videos, design professional-looking advertisements and produce fabricated customer testimonials to make an investment platform appear legitimate.
Some schemes promise unusually high returns within a short period.
Others use AI-generated videos of influential personalities to create the impression that the investment has been publicly endorsed.
The Securities and Exchange Commission (SEC) has warned about fraudulent investment schemes that use AI-generated voices, faces and videos to impersonate trusted public figures.
The Commission has also cautioned the public about anonymous operators, unverifiable social media accounts and unrealistic investment returns.
Securities and Exchange Commission Ghana
For someone hoping to grow their savings, the temptation can be considerable.
A person may see what appears to be an endorsement by a respected public figure, read convincing testimonials and decide to invest before conducting independent checks.
The consequences can be devastating, particularly when victims invest their savings or borrow money in anticipation of high returns.
What Ghanaians should do: Check the investment company's licensing status with the SEC before transferring money. Do not rely on screenshots of profits, testimonials or celebrity endorsements. Guaranteed high returns with little or no risk should be treated as a serious warning sign.
5. Impersonating businesses, employers and government institutions
Email Security for Small Businesses in Wilmington NC
AI can also help criminals impersonate organisations and their representatives.
A fraudster may create a convincing email account, imitate a company's communication style or generate a fraudulent invoice that looks similar to one routinely issued by a legitimate supplier.
For example, a business owner may receive an email claiming that a supplier has changed its bank account details and requesting that future payments be made to a new account.
An employee responsible for processing payments may act on the instructions without realising that the email is fraudulent.
Similarly, jobseekers may receive professionally written recruitment messages purporting to come from well-known companies or government agencies.
They may then be asked to pay application, processing or placement fees.
The danger is that AI-generated communication can make fraudulent instructions appear more consistent with an organisation's usual correspondence.
What businesses should do: Verify changes to supplier payment details through an independently established telephone contact. Introduce internal approval procedures for financial transfers and train employees to recognise impersonation attempts.
Jobseekers should verify vacancies through official recruitment channels and be suspicious of demands for unofficial payments.
6. Creating fake customer service agents and online business accounts
How AI Chatbots on WhatsApp Are Changing Customer Service
Another potential route is the impersonation of customer service representatives.
A fraudster may create a social media account bearing the name and logo of a bank, mobile money provider, online retailer or telecommunications company.
When a customer complains publicly about a failed transaction, the fake representative may contact them privately and offer assistance.
The victim could be asked to provide an OTP, click a link, download an application or share confidential account information.
AI can make these interactions more persuasive by helping scammers produce convincing responses and maintain conversations with multiple victims.
The victim believes they are speaking to someone who can resolve their problem, when the person may actually be attempting to gain access to their account.
This is particularly relevant to consumers who seek assistance through social media rather than official customer service channels.
What Ghanaians should do: Never share an OTP, PIN or password with anyone claiming to be customer support. Contact the company using the customer service details published on its official website or application.
Why AI scams are becoming harder to identify
The growing threat is not simply that criminals have access to more sophisticated technology. It is that AI can strengthen several stages of a traditional scam.
Consider the differences:
Traditional scam AI-assisted scam
Poorly written messages Professionally written phishing messages
Human voice impersonation Artificially cloned voices
Basic manipulated photographs Realistic AI-generated images
Manually produced fake advertisements AI-generated videos and promotional content
Limited personalised communication Tailored messages for different victims
Manual conversations with victims AI-assisted responses at greater scale
These comparisons describe potential capabilities, not proof that every scammer is using all these technologies.
AI does not need to be technically advanced to make a difference. Even relatively simple tools can help criminals create more convincing messages and advertisements.
There is also an important distinction between using AI to create fraudulent content and using AI to gain access to someone's financial account. In many cases, the technology helps establish trust, while the actual theft still depends on tricking the victim into transferring money or disclosing confidential information.
15. How Ghanaians can protect themselves against AI-powered scams
FEBRABAN TECH – ponto de encontro do mercado financeiro
As AI technology becomes more accessible, cybersecurity experts say individuals must become more deliberate about protecting their digital identities and financial information.
The Cyber Security Authority has repeatedly urged Ghanaians to verify suspicious requests, protect authentication credentials and report cyber incidents promptly.
The following measures can help reduce exposure to AI-assisted fraud.
1. Establish a family verification code
Families can agree on a private word or phrase to use when someone makes an unexpected request for money.
If a person claims to be a relative in distress, ask them to confirm the agreed phrase. Do not disclose the phrase in advance.
2. Always verify urgent financial requests
Scammers frequently create a sense of urgency to prevent victims from thinking carefully.
Whether the request comes through a telephone call, WhatsApp message, voice note or video, independently contact the person before sending money.
3. Never disclose your mobile money PIN or OTP
Your PIN and one-time password are confidential.
The Bank of Ghana has reiterated that legitimate banks, mobile money operators and payment service providers will not request customers' PINs, passwords or OTPs through telephone calls, text messages, emails or social media.
4. Activate two-factor authentication
Where available, enable two-factor authentication on your email, social media, banking and other important accounts.
This provides an additional layer of protection beyond a password.
5. Secure your WhatsApp account
Enable two-step verification, protect your email account and regularly review linked devices.
If you receive an unexpected request to link your WhatsApp account to another device, do not approve it unless you initiated the process.
6. Be careful about what you publish online
Public videos, voice recordings and photographs can provide material for impersonation.
Think carefully before sharing personal information, family details, travel plans and sensitive recordings publicly.
7. Verify investment opportunities independently
Before investing, check the identity and regulatory status of the company offering the opportunity.
Do not rely on advertisements, screenshots of profits, social media followers or endorsements by public figures.
8. Keep your phone and applications updated
Install security updates from official sources. Avoid downloading unknown applications or opening suspicious attachments, even when they appear to come from someone you know.
9. Monitor your financial transactions
Regularly check your mobile money and bank transaction notifications. Report unfamiliar transactions immediately.
10. Do not allow pressure to replace judgment
A genuine emergency can still be verified. A caller who refuses to let you independently confirm their identity should raise suspicion.
16. What should you do if an AI scammer has already stolen your money?
Shocked man in business setting holding smartphone and credit card, experiencing online fraud or banking issue
Discovering that you have been deceived can be distressing, particularly when the money involved represents savings, school fees, business capital or funds intended for an emergency.
However, victims should act quickly rather than remain silent out of embarrassment.
The following steps are important:
Step 1: Contact your bank or mobile money provider immediately.
Report the fraudulent transaction and request urgent assistance. Provide the transaction reference, recipient details, amount and time of transfer.
Ask whether the transaction can be stopped, traced or the receiving account restricted. Recovery is not guaranteed, but early reporting may improve the possibility of intervention.
Step 2: Report the incident to the Cyber Security Authority.
Ghana's CSA operates a 24-hour cybercrime reporting service.
Call or text: 292, 24-hour reporting point
WhatsApp: 0501603111
Email: report@csa.gov.gh
You can also submit an incident through the official CSA reporting portal.
Step 3: Report the matter to the Ghana Police Service.
Make a formal complaint and provide relevant evidence, including telephone numbers, social media accounts, transaction records and communications with the suspected fraudster.
Step 4: Preserve all available evidence.
Do not immediately delete conversations, voice notes, emails, screenshots or payment notifications.
These may assist investigators in identifying the individuals involved.
Step 5: Secure your accounts.
Change compromised passwords, revoke suspicious account sessions, enable additional authentication and contact your financial institution if your banking information may have been exposed.
Victims should also be cautious of individuals who subsequently contact them promising to recover stolen money for a fee. Recovery scams can target people who have already lost money.
17. The responsibility does not rest with individuals alone
While consumers have an important role to play, protecting Ghana's digital economy cannot be left entirely to individuals.
Financial institutions, telecommunications companies, social media platforms, technology providers and regulators must also respond to the changing threat.
Banks and mobile money operators need to strengthen systems for identifying unusual transactions, detecting suspicious account activity and responding quickly to customer complaints.
Telecommunications companies must continue working to address SIM-related fraud and impersonation.
Social media platforms also have a responsibility to improve the identification and removal of fraudulent advertisements, fake accounts and manipulated content that impersonates public figures and legitimate organisations.
For financial institutions, the challenge extends beyond detecting suspicious transactions. They must also educate customers about emerging methods of deception and make it easier to report suspected fraud.
The Bank of Ghana's October 2026 public advisory, titled Digital Fraud and the Cost of Ignorance, reinforced the importance of safeguarding financial credentials, avoiding suspicious links and verifying investment opportunities.
The CSA has similarly called for stronger collaboration between regulators, financial institutions, telecommunications operators, law enforcement agencies and the public.
Its position is that cybersecurity must be treated as an essential part of digital financial services rather than an issue to address only after money has been lost.
There is also a need for sustained public education in local languages so that people with different levels of digital literacy can understand the risks.
A person does not need to be highly educated or technologically sophisticated to become a victim of a convincing scam.
The more accessible AI becomes, the more important it is for public education to keep pace.
Conclusion:
In the age of AI, can you still trust what you hear and see?
Artificial intelligence is creating enormous opportunities for Ghana.
It can help businesses become more productive, improve education, support healthcare delivery and open new possibilities for innovation.
But the same technology can be exploited by criminals seeking to manipulate people and steal their money.
For Ghanaians, the challenge is becoming increasingly personal.
A familiar voice may not belong to the person you think it does. A video showing a respected personality promoting an investment may be fabricated. A professionally written message may be designed to steal your banking credentials.
The traditional habit of trusting what we see and hear is no longer sufficient in an environment where digital content can be manipulated.
This does not mean every telephone call, online advertisement or video should be treated as fraudulent. It means that financial decisions must be based on independent verification rather than appearance, familiarity or emotion.
Ghana's digital economy will continue to grow, but public confidence in mobile money, banking and online transactions will depend on how effectively the country protects its citizens.
The next time someone calls asking for money, remember: a familiar voice is not proof of identity, a convincing video is not proof of truth, and a professional-looking investment advertisement is not proof of legitimacy.
In the age of artificial intelligence, the safest response may be simple: stop, question and verify before you transfer.