How easy is it to create a fake AI video of someone you know? Apparently, pretty easy.
Last week, I told you about the time scammers tried to convince me I had two federal warrants. Welcome to week two of “I work in IT, cybersecurity and marketing, and apparently I have stories.”
This week’s story starts with my husband. Mind you, he also works in the tech world, so AI isn’t new to him.
He took one picture of me and used AI to turn it into a full-on moving video. Suddenly, there I was, moving, talking and doing things that never actually happened.
If you know me, you’ll notice quickly that something isn’t right. My dog, Mileaux, DEFINITELY doesn’t look like that. I also can’t remember the last time my arm didn’t have tattoos on it, and some of those tattoos are not where they belong! 😂
But honestly, that’s not the part that got my attention.
It was how easy it was to make.
My husband wasn’t trying to impersonate me, convince my boss to transfer money or pretend to be me to one of our customers. He was just playing with AI.
That’s when my cybersecurity brain kicked in.
If someone can do this for fun, what happens when someone does it with a completely different intention?
You Don’t Have to Be Famous to Be Deepfaked
When we hear the word “deepfake,” it’s easy to think about celebrities, politicians and viral videos. But think about how much material the average person already has online, especially business owners, executives, salespeople and, hello, marketers.
We post headshots, Facebook photos, LinkedIn posts, videos, reels, podcasts, webinars, speaking engagements, employee spotlights, company events, family pictures and pictures with our dogs.
I’m a marketer. Putting myself and the people I work with online is literally part of my job. I’m certainly not suggesting businesses stop doing that, but we do need to understand that something has changed.
The pictures, videos and audio we put online aren’t only content anymore. They can also become source material.
The FBI warns that generative AI has made it easier to create convincing synthetic audio, video, images and text, and that these tools are increasingly accessible to everyday users. My husband’s experiment made that very real to me. He didn’t need advanced technical skills or expensive equipment. He had a picture and an AI tool.
What Happens When an AI Deepfake Looks Real?
A few weeks before my husband made AI Heather, something much less funny happened to Columbia, Missouri, Mayor Barbara Buffaloe.
In August 2026, a resident contacted Buffaloe after seeing a video online that looked similar to one the mayor had previously posted. Except this version showed her saying vulgar things she never said.
The video was fake, but it was convincing enough that a resident reached out to her about it.
That’s the part that stuck with me. Someone watching the video wasn’t immediately sure whether it was real. Buffaloe later encouraged residents to use official sources when trying to verify information attributed to her.
Mine was funny. The mayor’s wasn’t. But both made me think about the same thing:
What happens when someone can make it look like you said something you never said?
For a public official, that can mean spreading false information. For a business owner or executive, the risks look different. Someone could appear to approve a decision. A voice message could appear to come from an executive. A fake video could damage someone’s reputation. Or an employee could receive a message that looks and sounds like someone they know, asking for sensitive information.
My Fake Video Wasn’t Perfect. That’s Important.
If you know me, you’d spot this one right away.
There are still clues that AI-generated content isn’t real. The FBI recommends watching for things like unnatural movements, visual distortions, inconsistent facial features, strange lighting or shadows, odd-looking accessories and unusual voice patterns.
So yes, pay attention to those things.
But I don’t want a business’s cybersecurity strategy to be:
“Don’t worry. We’ll be able to tell.”
Because what looks obviously fake today may not look obviously fake tomorrow.
Instead of only asking, “Does this look real?” maybe the better question is:
“Just because it looks real, does that mean it is?”
That’s the uncomfortable part of AI-generated content. We’re used to treating photos and videos as proof.
A Familiar Face Isn’t the Same as a Real One
For years, we’ve been taught to look for clues that something isn’t real. Check the sender. Listen to the voice. Look for strange wording or something that feels off.
Those are still good habits. But my little AI experiment made me realize we’re entering a world where spotting the fake may become harder than creating it.
That doesn’t mean we panic. It means we get a little more skeptical of what shows up on our screens.
If something seems unusual, especially when money, sensitive information or someone’s reputation is involved, don’t let a familiar face or voice be the only reason you believe it.
When the Joke Stops Being Funny
My husband’s video was funny. We laughed about how I sounded. We laughed about poor Mileaux. Then the funny part, wasn’t so funny..
Because all I could think was:
That was way too easy.
The scary part isn’t that AI can make a funny video of me. It’s that the same technology can be used to create something specifically designed to make another person trust what they’re seeing or hearing.
And that brings us right back to last week’s fake warrant call.
Those scammers didn’t need AI. They used fear, authority, personal information and a believable story. Now imagine adding a familiar face or voice to that same equation.
The scam doesn’t have to be perfect. It only has to be believable long enough for someone to act.
So, if there’s one thing I want businesses to take away from this, it’s that seeing isn’t necessarily believing anymore. Neither is hearing.
Because apparently my husband can make AI Heather now.
And AI Heather has absolutely NOT been authorized to approve wire transfers.
Diverse CTI believes the best defense against emerging threats like deepfakes starts with awareness. AI will continue to change what’s possible, and businesses need to change with it. Understanding the risks, educating employees and staying prepared can make the difference when something that looks real isn’t.