SLIDE 2 — Introduction

What is the “Illusion of Truth”?

Our brain tends to believe something simply because it looks real.

Repetition makes information feel true, even if it’s false.

Why this matters today:

Artificial Intelligence creates ultra-realistic fake videos, photos, and voices.

It is harder than ever to know what is real.

SLIDE 3 — What Are Deepfakes?

Definition:

Deepfakes = Fake images, videos, or voices created using Artificial Intelligence (deep learning).

They can replace:

Someone’s face

Someone’s voice

Entire events or scenes

Examples:

Fake Obama speech (Jordan Peele, 2018)

Fake Tom Cruise TikTok videos

SLIDE 4 — Why Deepfakes Fool the Brain

We trust what we see more than what we read.

If the face looks familiar, we believe it.

Seeing a video many times → becomes “true” in our mind.

We trust information more when shared by a celebrity or verified account.

SLIDE 5 — Real Incident #1: Fake Pentagon Explosion (2023)

AI-generated image showing an “explosion” near the Pentagon.

Shared by verified accounts → including Elon Musk.

Stock market briefly dropped.

Later confirmed FAKE by CNN, Reuters, and the U.S. Department of Defense.

Why this matters:

Even trusted people can spread fake AI content.

SLIDE 6 — Real Incident #2: Pope Francis in Balenciaga Coat (2023)

AI image of the Pope wearing a white designer jacket.

Millions believed it.

Shared globally before fact-checkers corrected it.

Shows:

Deepfakes can go viral even when harmless.

SLIDE 7 — Real Incident #3: Fake Zelenskyy Surrender Speech (2022)

Deepfake video showed President Zelenskyy telling soldiers to surrender.

Posted during the war in Ukraine.

Quickly debunked by BBC, Forbes, and several governments.

Shows:

Deepfakes can influence war, politics, and public opinion.

SLIDE 8 — Real Incident #4: AI Voice Scams

Criminals clone voices (e.g., Joe Rogan, parents, CEOs).

Used in scams:

“Mom, I’m kidnapped — send money!”

Victims believe it because the voice sounds real.

Shows:

Deepfakes are now used in cybercrime.

SLIDE 9 — Deepfakes in Politics

Fake videos of politicians saying things they never said.

Used to influence elections and public opinion.

Hard to detect once they spread.

Risk:

Democracy can be manipulated.

SLIDE 10 — Why Social Media Makes It Worse

Algorithms push shocking content.

Fake videos spread faster than real news.

Many people don’t check sources.

Verified accounts sometimes repost before verifying.

SLIDE 11 — The Illusion of Truth Effect

Key idea:

“If we see something many times, we believe it.”

Deepfakes abuse this psychological weakness by:

Looking extremely real

Being shared repeatedly

Triggering strong emotions

SLIDE 12 — The Dangers of Deepfakes

Fake scandals can destroy reputations

Misinformation can affect elections

Financial markets can react to fake events

AI voice scams are increasing

90% of deepfakes online = non-consensual pornography (MIT)

SLIDE 13 — How to Detect Deepfakes

Strange blinking

Unnatural movements

Unmatched lip-sync

Blurry or “melting” details

Impossible reflections or shadows

Is it from a trusted news outlet?

Has it been confirmed by multiple sources?

Reverse image search

Fact-check websites

SLIDE 14 — Solutions & Prevention

Raise awareness (education, digital literacy)

Governments work on deepfake laws

Social media platforms detect & flag AI content

AI watermarks (Google/Midjourney)

Personal responsibility: verify before sharing

SLIDE 15 — Conclusion

Deepfakes create the perfect illusion of truth.

We can no longer believe everything we see online.

To stay safe, we must:

Think critically

Verify sources

Question viral content

Understand how AI manipulates our perception

Final message:

→ Seeing is no longer believing.

SLIDE 16 — References

BBC News

CNN

Reuters

The Guardian

MIT Technology Review

Forbes

Washington Post