Defense Against The Deepfake

~20,000 founders and comms leaders start their week with The Brief.
Join them to get insights into the media, communications, and PR industry.

A fake video of your CEO hits the internet. 

It looks real. Sounds real. Makes the internet do what it does best: lose its collective ever-loving mind and clock 14 different reactions before anyone has confirmed what actually happened.

Your comms team scrambles.

Is it fake? Who verifies it? Who needs to know? What do we say? Why is everyone in the company-wide chat asking whether the CEO really just announced moving the company to Bora Bora? 

A deepfake is manipulated audio, video, or imagery that seems real but has been modified with AI. 

And while it may feel like the starting point of a crisis, it actually kicks off before that—with whether people know who to trust, where to go for the truth, and what to do when something definitely seems off. 

Deepfakes create a uniquely annoying PR problem: if convincing enough, saying “That’s not real” isn’t always gonna cut it. 

If we’ve learned anything from the internet, it's that people will absolutely believe a screenshot of a tweet from an account called @TotallyTheRealZendaya. 🙄

So what gives an organization credibility when seeing is no longer believing? 

Companies that have already established credible voices, recognizable channels, and strong relationships with their audiences have something to lean on when the internet starts serving up total BS. 

Take Airbnb. 

The brand and CEO have spent years building its business around trust and community. Its entire business model basically asks, “Hey, want to stay in a stranger's house?” and then somehow convinced millions of people that this was a totally normal Tuesday. 

Airbnb has described trust as the foundation of its business, and in a deepfake scenario, a reputation like this matters. Think of it as the girl-next-door effect—Airbnb is familiar, consistent, and generally recognizable in its voice and personality. 

If a video suddenly appeared of an Airbnb executive announcing something wildly out of character, the reaction might not immediately be “Welp. Guess Airbnb is at it again.”

It might be “Wait…that doesn't sound like them. Are we sure that’s real?” 🤨

It’s not a foolproof defense. But a strong, positive rep gives people a reason to pause before they hit Share. 

It just buys you time. 

Your response shouldn't begin with your public statement, but with your internal alarm system (immediately thought of 101 Dalmatians, just us?).

If Disney is ick to you, pretend we didn’t write the last sentence and put:

Otherwise, who’s gonna tell comms the deepfake exists?! 

Your social team may not be the first to catch it. Not even a reporter.

It might be someone named Bill from Finance scrolling X with his Raisin Bran at 7:45 a.m., and all of a sudden, that convincing CEO video comes up. 

Bill sends it to his manager.

His manager drops it in a work chat.

Someone sends it to Legal.

Someone else posts in the company-wide channel: “Is this real???”

The CEO gets an email from a reporter.

Customer service is blowin’ uppppp with questions.

And somewhere around 8:30, comms finally hears about the whole thing.

Cool. Cool cool cool. 

One fake video has turned into a full-company communications scavenger hunt (and unless there are eggs filled with candy, we’re not into it).

Your employees aren't just an audience waiting for the company's official response; they're potentially the first people to spot the problem.

Which means your internal comms strategy needs an answer to a very uncomfy but very important question: If someone—anyone—sees something suspicious, who do they tell?

Employees don't need to become AI experts. They just need to know what to do when something seems fishy, since they could be the first to encounter: 

The answer isn't to turn 10,000 employees into amateur forensic investigators. Don’t make “Deepfake Detection 101” the next Lunch & Learn or mandatory training.

No one needs another 42-minute module followed by asinine quiz questions.

The goal is much simpler: Give people a clear path from “Uh…is this real?” to “I know exactly what to do.”

That could be as simple as: see something suspicious ➡️

Don't share it ➡️

Capture the source ➡️

Report it ➡️

Verify through an official channel.

And the important part is that employees should already know 1) where to report it, 2) who handles it, and 3) where they can find the official answer. 

We want to avoid the nebulous “sOmEoNe ShOuLd TeLL HR” while everyone is actually internally screaming “NOT IT” 🫣 and assuming someone else will go ahead and relay the info. 

Then it’s crickets. 

Then disaster.

Worried you need to create an entirely separate AI Crisis Plan™ with its own flowchart, committee, chat group, and onboarding document (that everyone forgets exists until the exact moment they need it)?

Please don’t.

Deepfakes are new, but crisis communications isn't (this is your bread and butter, y’all!).

Simply fold synthetic media into the crisis protocols you already have. 

Ask the questions you're already supposed to be asking.

Who verifies?

Who gets alerted?

Who activates the crisis team?

Who communicates with employees?

Who talks to the media?

Where does everyone go for the official information?

And perhaps most importantly, how quickly can we move from “Is this real?” to “Here's what we know”?

PR Daily has a great prep plan for this kind of scenario: establish roles, create verified channels, draft a holding statement, and test 👏 the 👏 plan 👏 before 👏 you 👏 need 👏 it.

The unsexy truth about deepfake defense? Having the fanciest-schmanciest AI detection tool isn’t what you need.

You need people who trust your brand, know where the real information lives, and aren't going to forward a suss CEO video to 4,000 coworkers before breakfast.

That's the defense. And it starts long before the deepfake does.

So build the trust. Establish the channels. Tell people what to do.

Then, when reality gets a little AI-ish, you're ready.

Ash & Lizzy

Who's going where and why it matters. Not just job shifts - power dynamics, layoffs, and who's headed out.

👀 Business Insider has completely revamped its homepage experience to bring written, video, and audio coverage together in one place.

💔 No Jitter will stop publishing on September 30, 2026.

🧳 Jacob Bunge left The Wall Street Journal (after nearly two decades) to join the team at Longview Asset Management.

🎉 Dawn Chmielewski is now an Editor at Bloomberg’s LA bureau.

🎉 Alex Eichenstein will join The Information later this month to cover tech capital markets.

🎉 Luke Garratt is now Supervising Editor, Digital Video at The Associated Press.

🎉 Beth Healy is returning to Boston Globe Media as a Business Editor.

💔 Alyssa Lukpat has left The Wall Street Journal.

🎉 Gursimran Kaur Mehar has joined Bloomberg News as a Breaking News Editor.

🎉 Cristiano Lima-Strong is now the Senior AI & Government reporter for Bloomberg Government, covering AI policy across the White House, Congress, states, and more.

🧳 Megan Morrone has left her post at Axios. Stay tuned to her LinkedIn for the announcement of her next move.

💔 Michael Nuñez has left his post as Editorial Director at VentureBeat.

Get to know Ashley and Lizzy, learn more about The Colab, or pass The Brief along to a friend who should be reading this too.