Have you heard of the movie? Mountain Head still? It skipped the cinema and instead premiered at Max on May 31st.
Around four high-tech billionaires were convened in the gorgeous ski lodge Mountain Head (Named after Ayn Rand's novel, Fountain Head) For their annual acquisition, when the world around them falls into chaos and violence, they remain isolated, largely thanks to hyperreal deepfake AI videos that have made their software freely available to the general public and made them freely available.
Once AI video generation software was released, people quickly began to use it in the worst possible way, causing looting, riots, gang wars, and the nation's entire economy to collapse.
We've heard this is going on in the background, but through news reports it appears that the Tech Brothers are more interested in people who have the largest number in their bank accounts than the apocalyptic world after they guide them.
Character of Mountain Head Obviously a parody of today's technology leaders, Musk, Bezos, Altman and Zuckerberg, but not Silicon Valley's finest comedy roast; Mountain Head It absorbs faces from more flamethrowers, attacks the burnt earth, ashes, leaving only ashes.
Does that happen here?
Mountain Head It is one of the best Max movies at the moment, and the film currently sits at a very respectable 78% rating on Rotten Tomatoes.
Given the availability of platforms like Sora and Veo 3 that can generate realistic AI videos, I began to wonder how likely it is. Mountain Head Nightmare scenarios can actually happen in real life.
To rest my mind, I spoke with AI deepfake experts.
He didn't rest my mind. Instead, I'm beginning to wonder how much remains.
Creating a Chaos Burger
“You don't need to be refined to create chaos,” said Ari Abelson, co-founder of Operinorigins, which specializes in proofing images and videos. “Even the Slowfake campaign, and even the slowfake campaign, it makes Nancy Pelosi's video seem slower, creating moments of uncertainty that look like she's hiding.”
Of course, since the release of fake viral Nancy Pelosi videos, we have seen the emergence of powerful mass market AI video creation tools, such as Openai's SORA and Google's VEO 3.
“VEO 3 and other models offer a glimpse into the misinformation of a new generation, which is no longer 90% of viewers, as it is already indistinguishable from real human content for 90% of viewers.
“What you need now is to prove that the content is real because it's being created. If you're a company that relies on human media, the name of the game will prove that it's realistic and reliable when it's captured.”
For example, news companies need to rethink how they collect stories from sources, insurers need to secure claims, and the military needs to know which sources provide real evidence.
“Beyond this, we have to secure an archive of human content, so we have a tampered, immutable record of authenticity and we will not lose collective history with the enormous wave of AI slops,” Ableson says.
Do you think you weren't fooled by AI videos? Check out this video from Oxford University:
Inflection point
We seem to be at a horrifying inflection point where any form of communication through online sources can evaporate into global disruption if we can't reliably determine what is true from the fake.
“If this is not resolved, we enter a new phase where information sharing, history, news and facts are determined by opinions rather than what is actually happening,” Ableson says. “We may live in the future of AI control.”
This makes me more worried than before, so I asked Abelson if there was a way out of this mess. Luckily, he has an idea, and they are radical:
“The Internet needs to be diverged into the human and agent's internet. This is a fundamental change in the underlying infrastructure. The shared sense of reality in society is too important to leave it to probabilistic algorithms and unreliable heuristics.
A two-tier internet that contains only human content in one layer is an amazing idea I've never thought of before.
If possible, it could be revolutionary and our only defense against AI deepfakes that could destroy our trust in what we see online.
