AI Deepfakes Are Spiraling Out of Control — And No One Is Stopping Them

If you’ve spent any time online lately, you’ve probably seen something that looks very real – but it isn’t. That’s the power of AI Deepfakes, and right now, they’re spinning out of control. This is no longer a silly internet joke, it’s a real threat that’s everywhere.

Source:GAO 20-379SP, via University of Miami, 2026.

For normal people, it could reflect a serious consequence. For individuals, if a stranger uses AI to swap someone’s face into obscene videos, this could destroy person’s reputation and life overnight; For the public, scammers can use AI to imitate the voice and face of family members or friends, and steal large sums of money within minutes; For the society, the widespread spread of fake videos of politicians and celevrities stokes public fear and incites social divison. Once the lies spread, it is extremely difficult to explain clearly, because audiences always form their opinions first based on the first impression.

So what is the biggest issue? Anyone can do this, no technical skills required.

A few years back, this kind of tech was only for pros and big corporations. But now, anybody, on any device, can use free and easy AI tools to create a highly realistic fake video. That’s why fake content online has spiraled out of control.

Source: YouTube video “Deepfake Example. Original/Deepfake Elon Musk” by LipSynthesis, accessed March 2026.

What is even worse? Fake information spreads faster than truth. 

Platforms always tend to push shocking content, so deepfake videos can get millions of views in just a few hours. By the time people check the facts, it is already too late, many Internet users will never check the truth. As a Nature study shows that clear warnings about fake videos cannot stop people from believing them completely.

The true information spread speed vs the fake information spread speed: Snail vs Rocket

And what is the most powerless thing? We almost have no effective way to fight back. Even tools like MIL, which claim to detect fake videos, can’t keep up with how fast AI is getting better at making fakes. A 2024 study in Heliyon also points out that media and information literacy alone is not enough to fight disinformation, that means we still need better laws and faster tools.

In short, AI Deepfakes are out of control because they are easy to make, spread, and nobody could really manage them well. We can no longer ignore these issues, in this Internet age, people depend on platforms and media highly, and the governments and platforms should take stricter regulatory responsibility to protect digital trust and public safety . If it keeps going, the trust we place in online media and in each other will collapse entirely.