AI Deepfakes: When Seeing and Hearing Aren’t Believing Anymore

AI Deepfakes

When Seeing and Hearing Aren’t Believing Anymore

For most of our lives, we had a pretty simple rule:

If you saw it with your own eyes or heard it with your own ears, it was probably real.

A photograph was evidence.

A video was even better.

And if you heard the unmistakable voice of your daughter, grandson, friend or boss on the telephone, you had little reason to wonder whether it was actually them.

Artificial intelligence has changed that.

Today, technology can create convincing voices, photographs and videos of people saying or doing things they never actually said or did.

They're often called deepfakes.

And while some are created for entertainment, others can be used to deceive, embarrass, manipulate—or steal.

The good news is that you don't need to become an artificial-intelligence expert to protect yourself.

You just need to change one habit:

Don't automatically believe something simply because you can see it or hear it.

What Exactly Is a Deepfake?

A deepfake is digitally created or manipulated audio, video or imagery designed to make something appear real when it isn't.

Artificial intelligence can be used to imitate someone's voice, alter a photograph, or make it appear that a person in a video is saying something they never said.

The technology continues to improve, which means spotting a fake simply by looking or listening is becoming more difficult.

That's why the safest strategy isn't:

Can I tell whether this is fake?

It's:

Can I independently verify whether this is real?

That's a much stronger question.

Voice Cloning May Be the One That Hits Closest to Home

Imagine receiving a call from your grandson.

He's crying.

There's been an accident.

He's in trouble.

He needs money immediately.

And the voice sounds exactly like him.

This is where AI becomes especially powerful for scammers.

The Federal Trade Commission has warned that scammers can use short audio clips—potentially obtained from videos posted online—to create voice clones and enhance family-emergency scams.

You may not be able to determine whether the voice is genuine by listening carefully.

So don't try.

Hang up and call your family member yourself.

Use the number you've always used.

If they don't answer, contact someone else who can confirm where they are.

Video Isn't Automatic Proof Either

Video feels convincing because we're accustomed to trusting what we see.

But AI-generated or manipulated video can make it appear that someone is speaking words they never said.

That matters well beyond celebrity videos circulating online.

Imagine receiving a video message that appears to show someone you know asking for money.

Or a person on a video call who appears to be a trusted business contact.

Or a video of a public figure promoting an investment.

Seeing a face should no longer be the end of your verification process.

It may only be the beginning.

Photographs Can Be Created Too

AI can create remarkably realistic images of people, places and events that never existed.

That doesn't mean every unusual photograph online is fake.

It means photographs alone shouldn't automatically settle an important question.

This becomes especially important when an image is being used to make you:

Send money.

Click a link.

Invest.

Donate.

Share personal information.

Become frightened.

Or immediately forward something to everyone you know.

Whenever an image creates a strong emotional reaction, give yourself a moment before acting on it.

Scammers Want Emotion Before Logic

The technology may be new.

The psychology isn't.

Scams still rely heavily on the same emotions they always have:

Fear.

Love.

Urgency.

Greed.

Sympathy.

Anger.

Curiosity.

The scammer doesn't necessarily need to fool you for an hour.

They may only need to fool you long enough to send money, reveal a password or click a link.

That's why one of your best protections is surprisingly low-tech:

Slow down.

Watch for the Emergency Story

Deepfake technology can make traditional emergency scams much more believable.

Someone you know may supposedly be:

In jail.

In the hospital.

Stranded overseas.

Involved in an accident.

Kidnapped.

Unable to access their money.

Desperate for help.

Then comes the request:

Send money now.

And often:

Don't tell anyone.

That combination—emergency + secrecy + money—should immediately make you stop.

Call the person yourself.

Call another family member.

Verify what happened independently.

A genuine emergency won't disappear because you spent five minutes checking.

Create a Family Verification Plan

Families can prepare for this before anything happens.

Choose a private family word or phrase that can be used during an unexpected emergency.

Make it something that isn't easily guessed from social media.

Then establish another rule:

Nobody sends emergency money until the situation has been independently verified.

A safe word can help, but don't rely on it alone. If something doesn't feel right, call the person using a trusted number or contact another family member.

Think of the safe word as another lock on the door—not the entire security system.

Be Careful What You Share Publicly

You don't need to disappear from the internet.

But it is worth thinking about how much information is publicly available.

Social media can reveal names of family members, birthdays, vacation plans, workplaces, schools, pets, relationships and other details.

Public videos may also contain samples of someone's voice.

That information can make an impersonation attempt more believable.

Review privacy settings occasionally and think before publicly sharing information that could help a stranger understand your family.

Your grandson's name, school, dog's name and vacation destination may feel harmless separately.

Put together, they can tell quite a story.

Deepfakes Can Be Used in Romance Scams

Online romance scams already rely on creating trust with someone the victim hasn't met—or hasn't spent much time with in person.

AI can potentially make those deceptions more convincing through fabricated photographs, messages, voices and video.

Don't let a photograph or brief video conversation erase other warning signs.

Be cautious if an online romantic interest repeatedly has reasons they cannot meet, quickly declares strong feelings, experiences repeated emergencies, asks for money, introduces an investment opportunity or pressures you to keep the relationship secret.

The technology may look sophisticated.

The request eventually tends to become remarkably simple:

Send money.

Be Especially Careful With Investment Videos

You may encounter videos online that appear to show celebrities, financial experts or well-known businesspeople recommending an investment.

Don't assume the endorsement is genuine because you can see and hear the person.

The FBI has warned that criminals are using generative AI to make fraud schemes more believable, including synthetic audio, video and identification materials.

Before investing, research the opportunity independently.

Don't use the contact information, links or phone numbers provided in the suspicious video itself.

And remember:

A famous face doesn't make an investment safe.

Don't Click Because Someone Looks Familiar

Suppose you receive a message that appears to come from someone you know.

There's a photograph.

Maybe a video.

Maybe even a voice recording.

Then you're asked to click a link.

Stop.

Contact that person another way.

Ask:

“Did you send this to me?”

It takes thirty seconds.

And thirty seconds is much cheaper than recovering a hacked account.

Look at the Whole Situation

People sometimes search for tiny technical clues in deepfake videos:

Does the mouth move strangely?

Do the eyes blink normally?

Does the lighting look wrong?

Are the hands unusual?

Those clues can sometimes help.

But technology changes quickly.

A better approach is to examine the story surrounding the content.

Ask:

Who sent this?

Why am I receiving it?

What do they want me to do?

Are they creating urgency?

Are they asking for money?

Do they want personal information?

Are they telling me to keep something secret?

Can I verify this another way?

You don't have to prove something is fake before deciding not to trust it.

Use the Two-Channel Rule

For anything important, verify through a second method.

If you receive a phone call, send a text using the number already in your contacts.

If you receive a text, call the person.

If you receive an email from your bank, open the bank's official app instead of clicking the email link.

If someone sends you a video asking for money, call them independently.

You're looking for confirmation from somewhere outside the suspicious communication.

Don't Let Embarrassment Keep You Quiet

Maybe you believed the call for a few minutes.

Maybe you clicked something.

Maybe you sent money.

Maybe the video looked completely real.

Don't let embarrassment stop you from asking for help.

These scams are deliberately designed to be convincing.

The sooner you act, the better your chances of limiting the damage.

If money was sent, contact the bank, credit-card company, payment service or other financial institution involved immediately.

Change compromised passwords.

Preserve messages, phone numbers, emails and transaction records.

Report fraud to the appropriate authorities.

Teach the Family Both Directions

This isn't only something older adults need to understand.

Children and grandchildren should know about deepfakes too.

Someone could impersonate you.

Your family might receive a message that sounds like you saying:

“I lost my phone.”

“I'm stranded.”

“I need money.”

“Don't call anyone.”

So have the conversation both ways.

Everyone in the family should know the verification rule.

No hurt feelings.

No embarrassment.

Just:

“Before we send money, we verify.”

The 30-Second Deepfake Check

When something unexpected arrives, ask yourself:

Is this unusual?

Is it emotional?

Is it urgent?

Is money involved?

Am I being asked to keep it secret?

Can I verify it another way?

If several of those questions raise concerns, stop.

Don't click.

Don't pay.

Don't provide information.

Verify first.

You Don't Have to Be Afraid of AI

Artificial intelligence can do many useful things.

It can help people learn, communicate, organize information and solve everyday problems.

The answer isn't to become suspicious of every photograph, video and telephone call.

It's simply to understand that technology has changed.

There was a time when:

“I heard it myself”

or

“I saw the video”

felt like enough proof.

Today, sometimes it isn't.

So we're adding one more step:

I saw it. I heard it. Now I'm going to verify it.

That's not paranoia.

That's good judgment in a world where technology can imitate reality better than it ever could before.

Seeing may still be believing.

But when money, safety or someone you love is involved—

checking is better.

The World U — Helping people over 60 live smarter, safer, and more confidently.