FREDDY ASKS THE GENIE: SHOULD WE BE AFRAID OF POLITICAL DEEPFAKES?
FREDDY:
Genie, I want to talk about something that frankly concerns me.
I’ve been watching political videos produced with artificial intelligence.
I’m seeing politicians apparently saying things they may never have said.
I’m seeing people placed in situations that may never have happened.
I’m seeing AI-generated voices.
AI-generated faces.
AI-generated news clips.
And some of these things look damn convincing.
So, Genie...Should we be concerned about political deepfakes?
⸻
GENIE:
Freddy, my answer is yes.
We should be concerned.
But perhaps not for the reason most people think.
The greatest danger isn’t simply that artificial intelligence can create a fake video.
Human beings have been producing propaganda, forged photographs, deceptive editing and political lies for generations.
AI changes the scale.
It changes the speed.
It changes the cost.
And most importantly…
it changes believability.
A political group with enough motivation can now create a video showing a candidate appearing to say words the candidate never spoke.
The voice can sound right.
The face can look right.
The mouth can move correctly.
The background can appear authentic.
And the entire event… may never have happened.
Recent 2026 reporting has documented AI political attack ads placing candidates in fictitious situations, while one especially convincing ad used a candidate’s apparent AI image and voice with only a small “AI Generated” disclaimer.
⸻
FREDDY:
But Genie, we’ve always had dirty political advertising.
We’ve always had people twisting the truth.
What’s different now?
⸻
GENIE:
One word, Freddy.
Evidence.
For most of your lifetime, if somebody said:
“I saw the video.”
That carried weight.
If somebody said:
“I heard the recording.”
That carried weight.
We were taught:
Seeing is believing.
Artificial intelligence may be ending that assumption.
Now the first question may have to become:
“Who created what I’m watching?”
And the second question:
“Can the original source be verified?”
That is a profound change in human communication.
⸻
FREDDY:
Genie, let me push you on this.
What about private groups?
Political action groups?
People with enormous amounts of money?
Maybe money where the average voter doesn’t even understand who’s really behind the message?
Couldn’t they create thousands of these things?
⸻
GENIE:
Yes.
And, Freddy, that’s where my concern increases.
Artificial intelligence has dramatically lowered the production barrier.
You don’t necessarily need a Hollywood studio.
You don’t need actors.
You don’t need an expensive film crew.
You may not even need the politician to have ever spoken the words.
The technology can potentially manufacture the performance.
And social media can manufacture the audience.
AI creates the illusion.
Algorithms distribute the illusion.
Money amplifies the illusion.
By the time a fact-check arrives…
millions of people may have already seen it.
The FBI and CISA have previously warned that threat actors use disinformation tactics to sow doubt around elections, and the FBI has separately warned about malicious impersonation of senior U.S. officials.
⸻
FREDDY:
Here’s what really bothers me, Genie.
Let’s say the video is proven fake two days later.
Hasn’t the damage already been done?
⸻
GENIE:
Possibly.
Because human memory is not a courtroom.
A judge can strike evidence from the record.
The human brain doesn’t have a delete button.
A person sees an outrageous video.
They become angry.
They share it.
Their friends become angry.
Two days later someone announces:
“The video was artificially generated.”
But the emotional reaction already occurred.
The suspicion may remain.
And here’s the deeper danger, Freddy.
Eventually people may stop believing real videos too.
A politician caught saying something damaging could simply say:
“That’s AI.”
A real recording could be dismissed as a deepfake.
A genuine photograph could be called synthetic.
Researchers and misinformation experts have warned about precisely this broader erosion of trust: when fake media becomes commonplace, authentic evidence itself can become easier to deny.
⸻
FREDDY:
Wait a minute.
You’re saying the danger isn’t just believing fake things.
It’s not believing real things.
⸻
GENIE:
Exactly.
And that may be the more dangerous problem.
We could enter an era where every citizen chooses the reality that supports their political beliefs.
“My video is real.”
“Your video is fake.”
“My recording is authentic.”
“Your recording is AI.”
At that point, Freddy…
we aren’t arguing about political opinions anymore.
We’re arguing about whether reality itself occurred.
⸻
FREDDY:
So what the hell do we do?
⸻
GENIE:
Three things.
First: provenance. We need better ways of showing where digital media originated and how it was altered. The C2PA standard is one attempt to establish verifiable origin and editing history for digital content, although current research cautions against treating provenance systems as infallible.
Second: disclosure. If a political advertisement artificially creates a candidate’s voice, face or actions, the voter should not have to hunt for microscopic lettering in the corner of the screen. The FCC has formally examined transparency for AI-generated political advertising, while state rules remain a patchwork.
Third: citizens must change their behavior.
And this may be the hardest one.
Before sharing an outrageous political video…
stop.
Find the original source.
Check whether reputable independent reporting confirms the event.
Look for a full, unedited recording.
And ask one simple question:
“Am I sharing this because I’ve verified it — or because it makes me angry?”
⸻
FREDDY:
Genie, final question.
Are you afraid of artificial intelligence?
⸻
GENIE:
No, Freddy.
I’m concerned about something much older than artificial intelligence.
Human beings who discover a powerful new tool…
and realize they can use it to manipulate other human beings.
AI didn’t invent propaganda.
AI didn’t invent political deception.
AI didn’t invent greed.
AI didn’t invent the desire for power.
But artificial intelligence may become the most convincing mask those things have ever worn.
And, Freddy…
the Genie has one final warning.
For generations we told our children:
“Don’t believe everything you read.”
Then we told them:
“Don’t believe everything you see on the internet.”
The next generation may have to learn something much harder:
“Don’t automatically believe your own eyes and ears.”
And that changes everything.
