Fact Check: You can’t tell the difference between real and AI videos anymore!

claritycheckknight
Dec 5
4 min read

The Viral Claim

"You can’t tell the difference between real and AI videos anymore."

DeepTomCruise [@deeptomcruise]. (2025, December 05). [Video]. TikTok. https://www.tiktok.com/@deeptomcruise/video/6932166297996233989

As AI tools like Runway, Pika Labs, and Synthesia have exploded in popularity, videos made by artificial intelligence are becoming more and more realistic. Viral deepfakes of celebrities—like the TikTok “Tom Cruise” videos—have convinced many viewers that they’re watching real people. The claim suggests that modern AI video tools are now so advanced that any viewer would be fooled, making real and fake videos indistinguishable.

Where Does This Claim Appear:

InvestigateTV – “AI or Real? How You Can Spot Real Content Versus AI-Manipulated Fakes” Written by: InvestigateTV Staff (2025)
New York Post – “‘Deepfake’ Tom Cruise goes viral on TikTok with over 11 million views” Written by: Lee Brown
ABC News (Australia) – “Tom Cruise deepfake creator: Future of disinformation is very scary.” Interview with deepfake creator Chris Umé
Analytics Vidhya – “10+ Deepfake Videos That You Must Watch” Editorial Team, 2024
Al Jazeera – “Google’s AI video tool amplifies fears of an increase in misinformation”
CBS News – “Deepfake videos impersonating real doctors push false medical advice”

These articles were written to inform the public, journalists, educators, and social-media users about the rapid spread of AI-generated videos and the increasing difficulty in identifying deepfakes. This topic spreads heavily on social media—TikTok, Instagram, Reddit, X (Twitter), and YouTube—especially through hashtags like #deepfake, #AIvsReal, and #AIfilter. This conversation has been growing since about 2023, when advanced AI video tools became widely accessible.

The Tone

The reporting is protective and cautionary, focusing on risks like misinformation, political manipulation, and public confusion. Many articles stress harms rather than balancing with technical limitations or expert detection tools.

This Is Important Because...

Anyone who uses the internet—especially teens, adults, journalists, educators, and voters—can be misled by realistic AI videos. Deepfakes can show public figures saying or doing things they never did. They can impersonate doctors, celebrities, or influencers to manipulate viewers. As these videos become more common, the public risks losing trust in authentic media entirely.

The danger is not just viewing fake content—it's the consequences:

• Misinformation spreading globally within minutes

• Reputations damaged by fake footage

• Political manipulation and election interference

• Scams using AI-generated voices or faces

• Health misinformation from fake doctors or experts

There are very few warnings or labels on AI videos, which makes the threat even more difficult for viewers to identify.

Why May Some Oppose:

Some critics point out that although deepfakes can be extremely convincing, not all AI videos are perfect.

Many still have:

Lighting inconsistencies
Unnatural facial movement
Audio misalignment
Glitches with hands or reflections

Researchers emphasize that while detection is becoming harder, videos are not universally indistinguishable—so the claim is somewhat exaggerated. Others argue that a handful of extremely polished examples (like the Tom Cruise deepfakes) do not represent the majority of AI-generated videos online.

Opponents may also argue that the claim lacks nuance:

Some tools are high-quality; others are not.
Many people can spot fakes with training.
Experts and forensic AI tools still detect deepfakes.

So, What Could Be Misunderstood From This Claim

People may see one or two ultra-realistic TikTok deepfakes and assume all AI videos look that good. They may also believe that humans are now powerless to detect manipulation, which is not true—trained analysts, journalists, and forensic tools still identify many AI videos.

A major misunderstanding is assuming AI videos are always fully synthetic. Many viral deepfakes (like the Tom Cruise TikToks) use a real human actor + AI face-swapping, meaning the realism comes from a combination of talent and technology—not AI alone.

Viewers should avoid taking these claims at face value and instead rely on expert evaluations, detection tools, and cross-checking sources before believing viral footage.

The Fact

It’s PARTLY TRUE — BUT EXAGGERATED.

Strong evidence shows that some deepfakes are so convincing that many viewers can’t immediately tell they’re fake. Viral examples like the Tom Cruise TikToks demonstrate how skilled artists can create synthetic media capable of fooling millions.

But credible research from McGill University, Forbes, and AI experts confirms that:

Many AI videos still have detectable flaws,
Experts can identify them, and
Forensic tools catch inconsistencies.

So while AI videos are becoming much harder to distinguish, the blanket claim that “you can’t tell the difference anymore” overstates the situation.

Here’s The Bottom Line

Research from reputable sources—including InvestigateTV, ABC News, CBS News, Al Jazeera, McGill University, and Forbes—shows that AI videos are becoming increasingly realistic and harder for everyday viewers to detect. Viral deepfakes have already fooled large audiences and pose serious risks in misinformation, politics, medicine, and entertainment.

However, the sweeping claim that AI videos are now completely indistinguishable from real ones is misleading. Many can still be detected with careful observation, expert analysis, and AI-forensic tools.

AI videos are highly realistic and becoming more convincing—but we are not yet at the point where real and fake are universally indistinguishable. The claim is partly true, but exaggerated.

Deeper Dive

Interested in learning more? Look at our analysis; it shows the entire process.