The New AI Scams Targeting Your Family

A person reviewing a social media photo grid on a phone and laptop.

A familiar voice used to feel like reassuring evidence. If someone sounded like your daughter, brother, or close friend, you had a reason to believe it was them. Artificial intelligence has made that assumption less dependable.

The Federal Trade Commission warns that a short recording can be used to imitate a loved one’s voice. A clip posted online may be enough to help a criminal produce a convincing impersonation. The voice can be familiar even when the story is false.

This does not mean every unexpected call is a scam. It means that a voice, photograph, or video should no longer settle an important decision on its own.

Old schemes have more convincing disguises

The underlying tricks are often familiar: a family emergency, a promising investment, a romantic connection, or an apparent authority figure asking for money. AI can make the presentation more believable and help criminals contact more people.

In its warning about AI-assisted financial fraud, the FBI describes polished messages, invented social profiles, synthetic voices, and fabricated videos. These can support a false identity or make a financial offer appear legitimate. A message without spelling mistakes is not necessarily a trustworthy message.

Some deceptions unfold over time. A person may exchange messages with an apparently friendly contact before an investment or payment request appears. Others rely on a sudden shock. The important connection is the request being made, not simply how professional the conversation looks.

The family emergency that demands secrecy

Imagine a fictional call in which a voice that sounds like your adult child says there has been an accident. Another person takes over, introduces themselves as a lawyer, and says a payment is needed immediately. You are told not to contact anyone else.

That combination of urgency, an apparent authority figure, and secrecy is described in the FTC’s guidance on fake emergencies. Its purpose is to leave little room for an independent check. A request for cryptocurrency, a wire transfer, or gift-card numbers adds another serious warning sign.

The strongest response does not require deciding whether the voice was generated by AI. It is a separate conversation with the real person, using a number already known to belong to them. If they cannot be reached, another trusted relative or friend may help establish what is happening. A new number supplied by the caller is not an independent route.

Familiar details are not proof

Knowing a family member’s name, workplace, or recent holiday can make an impersonation persuasive. Those details may also have appeared in public posts or earlier messages. Recognition and verification are different things.

The FBI suggests a private family word or phrase as another identity check. That can add protection, but it should not replace a callback through a trusted channel, particularly when money is involved. A secret that has been shared widely or exposed elsewhere is no longer a dependable safeguard.

Visible glitches or unnatural speech may raise suspicion. Their absence does not establish that a recording is genuine. Families should not have to become experts in spotting digital fakes before they are allowed to pause an unexpected request.

If someone has already sent money

The FTC recommends contacting the bank or payment company promptly, explaining that a scam occurred, and asking whether the payment can be reversed. Recovery is not guaranteed. Receipts, original messages, account names, and payment details can help with a report.

Fraud can be reported through the FTC’s ReportFraud service or the FBI’s Internet Crime Complaint Center. If account credentials were exposed, the affected account also needs attention through the provider’s official security and recovery process.

For a family, the response matters beyond the payment. Blame can make someone hesitate to mention the next suspicious message. A calm conversation gives everyone a better chance of interrupting a scheme early. The useful habit is simple: an unexpected request can wait long enough to be checked.

Sources checked November 6, 2026. AI assisted with drafting and source review. Read our full disclaimer.

Photograph: Plann / Unsplash. Illustrative image.

Leave a Comment

Your email address will not be published. Required fields are marked *