Imagine receiving a call from a family member asking for urgent help. The voice sounds right, the way they speak feels familiar, and the situation seems serious. There is just one problem: it may not actually be them on the other end of the line.
With artificial intelligence tools, criminals can create audio that mimics the voices of relatives, friends, coworkers, and even company executives. AI voice cloning has made emergency and impersonation scams increasingly convincing.
This threat is part of a broader evolution in online fraud. As we explain in our article on 5 trending digital scams and how AI is making fraud more dangerous, criminals are already using AI to improve phishing, fake call centers, and other forms of social engineering.
That makes recognizing suspicious calls and verifying a story before taking action more important than ever.
Voice cloning uses artificial intelligence to reproduce characteristics of a person’s speech, including tone, rhythm, and intonation.
It is closely related to deepfake technology, which can generate convincing fake images, videos, and audio. In our guide to deepfakes and how scammers are using them, we explain how these tools can be used for financial fraud and impersonation.
When used legitimately, voice cloning can have many applications. The problem begins when scammers use an artificial voice to impersonate someone you know and gain your trust.
A criminal might pretend to be:
The goal is usually to make you react before you have enough time to verify what is actually happening.
The script can vary, but these scams usually exploit two things: trust and urgency.
You receive a phone call or voice message that appears to come from someone you know. They describe an accident, financial problem, arrest, or another emergency and ask for money, personal information, or immediate action.
Scammers may combine the cloned voice with personal information found online or obtained through leaked data, making the story even more believable.
The Federal Trade Commission warns that scammers can use a short audio clip, potentially taken from content posted online, to clone a loved one’s voice and create a fake emergency. Its key advice is simple: do not trust the voice alone.
It is not always possible to detect fake audio simply by listening to it. As the technology improves, relying exclusively on imperfections in the voice becomes less reliable.
Still, certain warning signs should make you more cautious.
“I need the money right now.”
“Don’t tell anyone.”
“Send it before it’s too late.”
Creating pressure is a common scam tactic. The more urgent the situation feels, the less time the victim has to verify the story.
Be suspicious if someone you know suddenly asks for money, banking information, passwords, authentication codes, or other sensitive information.
Even if the voice sounds legitimate, the request itself should make sense.
A scammer may give you reasons not to hang up or contact anyone else.
That behavior is an important warning sign.
Unusual pauses, changes in intonation, or strange responses may appear in manipulated audio. But these signs should never be your only method of detection.
Even a convincing recording can be fake.
Keep reading — PSafe
https://www.psafe.com/en/blog/5-trending-digital-scams-how-ai-is-making-fraud-more-dangerous/ https://www.psafe.com/en/blog/deepfakes-the-new-weapon-of-digital-scammers/ https://www.psafe.com/en/blog/protect-social-engineering-attacks/
The most important rule is simple: end the call and verify the story using a phone number you already know.
If someone calls claiming to be your child, for example, hang up and call the number saved in your contacts.
Do not automatically call a number provided during the suspicious conversation.
If you cannot reach the person, contact another trusted family member or friend who may be able to verify the situation.
The FTC’s guidance on family emergency scams gives similar advice: resist pressure to send money immediately and contact the person through a phone number you know is legitimate.
One simple precaution is to agree on a family safe word or verification question that only people close to you should know.
If someone contacts you claiming there is an emergency, this information can provide an additional layer of verification.
However, it should not replace other security measures. If something feels suspicious, end the conversation and contact the person through another trusted channel.
The same principle applies in the workplace.
A familiar voice should not be considered sufficient proof to authorize payments, change banking information, or share confidential data.
Companies can reduce the risk by creating verification procedures for sensitive requests, particularly financial transfers. Voice cloning is one part of a broader social-engineering problem. OIf you receive an unusual request attributed to a manager or coworker, verify it through another corporate channel before acting.
The technology criminals use may evolve, but several precautions remain effective:
Above all, remember that recognizing someone’s voice is no longer enough to verify their identity.
Social-engineering scams work because they exploit human emotions: fear, trust, concern, and urgency.
Artificial intelligence can make the approach more convincing, but it does not change one of the best defenses against fraud: verify before you act.
If a call or voice message contains an urgent request, end the contact. Reach the person through a phone number or communication channel you already trust and confirm the story directly.
A few minutes of verification may be enough to prevent a familiar voice from being used against you.
FONTES
https://consumer.ftc.gov/consumer-alerts/2023/03/scammers-use-ai-enhance-their-family-emergency-schemes https://consumer.ftc.gov/articles/scammers-use-fake-emergencies-steal-your-money
Google Pics began rolling out on September 1, 2026, with AI tools for creating and…
When something unusual happens on a smartphone, many people immediately blame the SIM card. But…
A phone overheating does not always mean something is broken. Gaming, camera use, charging, weak…
ChatGPT for Teens began rolling out globally on August 18, 2026. The experience is designed…
Google published the August 2026 Android security update bulletin on August 3, with patch levels…
Do you turn Wi-Fi off whenever you leave home because you expect your battery to…