Mark Zuckerberg's Koolau Ranch on the Hawaiian island of Kauai covers 930 acres. It includes two football-field-sized mansions, a gym, guest houses, tennis courts, a private water system, and an underground tunnel leading to a shelter with blast doors, water supplies, autonomous power generation systems, and an emergency exit. The estimated cost exceeds $200 million. It's not an isolated case. In Texas, la SAFE (Strategically Armored & Fortified Environments) is building a $300 million bunker complex that will house 625 people, with units selling for up to $20 million each. The Aerie Project It promises the super-rich "a robust defense against the most serious threats, including nuclear fallout and electromagnetic pulses." Tropical pools with simulated sunlight, gourmet restaurants, medical centers with artificial intelligence. A five-star apocalypse. Que pasa, amigos?
The Map of Billionaires' Hides
Reid Hoffman founder of LinkedIn, revealed to the New Yorker That at least 50% of Silicon Valley's super-rich already own what they call "apocalypse insurance." This isn't rhetoric. It's a strategy documented with names, places, and numbers.
Jeff Bezos founder of Amazon, purchased two $147 million mansions on Indian Creek Island in Florida. Larry Ellison, billionaire of Oracle infrastructures , bought (also) a property on the Hawaiian island of Lanai. Peter Thiel, Co-founder of PayPal, chose New Zealand. Jack Ma founder of Alibaba, director James Cameron and the financial guru William Foley They've all built post-apocalyptic bunkers in remote locations.
But let's get back to good old Mark for a moment. When asked if he was building a doomsday shelter, Zuckerberg responded with a firm "no." He called the approximately 465-square-foot underground space "a small shelter, like a basement." As if regular basements had blast doors, secret tunnels, and six-foot-high walls to hide the construction site from prying eyes.
All the workers involved in the project had to sign strict confidentiality agreements. "It's like Fight Club," one of them said. "We can't talk about it."
The Business of the Apocalypse
Companies that build bunkers for The super-rich are experiencing an unprecedented boom. Brian Cramden, President of Hardened Structures, a Virginia company specializing in multi-million dollar fortified homes and bomb shelters, he explained to CBC News that requests are increasing exponentially.
La SAFE The first complex is scheduled to open in 2026. The Aerie project will offer a network of luxury residential bunkers in 50 cities across the United States, with 1.000 affiliated locations planned worldwide. These aren't basic shelters. The modern bunkers include suites with wood finishes, designer furniture, and swimming pools with simulated solar lighting (like the one in Oppidum bunker in the Czech Republic), wine vaults, gyms, arcades, climbing walls and even bowling alleys.
Other features of high-end bunkers include air filtration systems to prevent contamination, wind turbines and solar panels to provide electricity, and dedicated spaces for growing plants and raising fish to provide almost unlimited food. Their battery life is claimed to be five years. The cost of a basic unit starts at $1,5 million.
What the super-rich really fear
Douglas Rushkoff, media theorist and author of the book Survival of the RichestHe recounted a surreal meeting with five of the world's richest people in an undisclosed location. They wanted advice on the future of technology. But the questions they asked him weren't about innovation or business. They were about survival.
“These billionaires they are building bunkers with moats "For protection," Rushkoff said. They asked him which location was better for survival: Alaska or New Zealand? How to maintain control of armed guards when money is no longer valuable? Some thought of disciplinary collars, others of electronically controlled food access codes, still others of the use of defense robots. One of them would enlist a dozen Navy SEALs ready to intervene "at the right moment."
In his book, Rushkoff explains that many billionaires, especially those in Silicon Valley, are preparing for what they call “the Event“. A euphemism for environmental collapse, social unrest, a nuclear explosion, a solar storm, an unstoppable virus, or a devastating cyber attack.
“They have the belief,” Rushkoff writes, “that with enough money and technology, rich men can live like gods, and transcend the calamities that befall everyone else.”
Even AI scientists want bunkers
If the super-rich fear the future, the scientists who are building that future are even more afraid. Ilya Sutskever, the genius behind ChatGPT and co-founder of OpenAI, told his researchers at a meeting in the summer of 2023: "Once we're all in the bunker..." A confused researcher interrupted to ask for clarification. The response was even more surprising: "We'll definitely build a bunker before releasing the AGI."
As reported by Karen Hao in his book Empire of AI: Dreams and Nightmares in Sam Altman's OpenAI, Sutskever's plan was to protect key scientists of OpenAI from the geopolitical chaos or violent competition between world powers that he predicted would explode after the release of artificial general intelligence. This wasn't a joke. It was a serious proposal.
Sutskever left OpenAI in May 2024 to found Safe Superintelligence, a company dedicated to developing secure artificial intelligence systems. “Our first product will be secure AI, and we won't do anything else until then,” he said. The company initially raised $1 billion, then another $2 billion, reaching $1 billion in 2019. TechCrunch a valuation of $32 billion as of April 2025. The fact that concerns about AI safety are being echoed among investors shows that it is not simply paranoia.
AGI: When artificial intelligence will surpass human intelligence
The AGI (Artificial General Intelligence) is artificial intelligence capable of tackling any human task with equal or greater efficiency. It's one of the AI topics we often return to. In recent years, we've provided you with the position of virtually all of Big Tech.
Sam altmanCEO OpenAI, said in December 2024 that it will arrive “sooner than most people in the world think.” Sir Demis Hassabis, Co-founder of DeepMind, predicts in the next five to ten years. Dario Amodei founder of anthropic, wrote that his preferred version of the term, “powerful AI,” could be with us as early as 2026.
According to the 2025 AI Index Report with the Stanford Institute for Human-Centered AIAI models have already achieved superior performance to humans in complex tasks such as natural language understanding, visual recognition, and advanced mathematical problem solving. "The models are improving rapidly, reaching (and in some cases surpassing) human capabilities," the study states.
DeepMind identified four main risk categories associated with the development of AGI: improper use (an attacker could exploit an AGI system to carry out cyber attacks or develop viruses for military purposes), misalignment (AI may pursue goals other than those programmed), errors (unpredictable malfunctions) and structural risks (systemic impacts on society).
Bottom line: The super-rich and “creators” are scared—should we be too?
Rushkoff told a revealing anecdote. “The founder of a successful social media platform told me he’s erasing all his traces on the internet. He doesn’t post anything negative about AI, for fear it might be used against him when it takes over one day.” Rushkoff replied that a truly intelligent AI could deduce what the billionaire deliberately withholds. “So the omniscient AI would find out anyway.” The billionaire couldn’t help but curse.
The paradox is stark: The people developing the world's most advanced artificial intelligence are also the most terrified of its consequences. They continue to push forward with research while designing escape routes. It's like building a dam knowing it will collapse, but instead of reinforcing it, you buy a boat.
Let's not make it all dramatic, though. Dame Wendy Hall, a professor of computer science at the University of Southampton, is skeptical of the apocalyptic predictions, especially regarding the timeframe: "The scientific community says AI technology is extraordinary, but it's far from human intelligence." Several "fundamental steps" would still be needed before we reach true AGI.
The Billionaires' Utopia
Elon Musk he surmised that superintelligent AI could usher in an era of "high universal income." He recently endorsed the idea that AI will become so cheap and widespread that virtually everyone will want their own "personal R2-D2 robot," to quote Star Wars.
“Everyone will have the best medical care, food, housing, transportation, and everything else. Sustainable abundance,” he wrote enthusiastically.
But if this optimistic vision were realistic, why are AI proponents themselves building bunkers? the super rich have accumulated $42 trillion over the last ten years while designing disaster shelters? The contradiction is stark.
Who remains outside the bunkers
The question that no one among the The super rich seem to ask themselves is simpleAnd the others? Rushkoff says he met a former bodyguard for a billionaire with his own private bunker. He asked him what the security plan was. The response was chilling: "The security team's first priority, if this were to actually happen, would be to eliminate the boss and take control of the bunker." And he didn't seem to be joking.
Bunkers, however luxurious, remain closed systems. Fragile. If the outside world becomes completely uninhabitable, no amount of technology can stop them. Hydroponic gardens with humidity sensors and computerized irrigation systems work well on farms, where dead plants can be replaced. In sealed bunkers, this isn't possible. They require external resources: maintenance. Spare parts. People.
The vast majority of citizens do not have access to a nuclear fallout bunker, nor even the means to build one. Relatively few people are allowed in government bunkers, intended for the continuity of state operations. According to a Oxfam International report, billionaires' wealth has increased by several trillions while global inequality has reached unprecedented levels.
The super-rich pay less than 0,5% of their wealth in taxes.
The Doomsday Clock on the Wrists of the Super-Rich
On January 28, 2025, the Bulletin's Science and Security Board revealed the time of the Doomsday Clock (Doomsday Clock). The SASB considered multiple global threats: nuclear weapons proliferation, disruptive technologies like artificial intelligence, the Russia-Ukraine war, the Israel-Hamas conflict, biological threats, and the climate crisis. The Doomsday Clock 2024 was stopped at 90 seconds to midnight. The closest it has ever been recorded.
For the super-rich, the clock strikes dinnertime. For the rest of the world, it strikes something else. Rushkoff calls this mentality “The Mindset”: a vision in which winning means isolating oneself from the damage one has caused, not solving it.It's the ideology that led Musk to Mars, Altman to invest in consciousness uploading, and Zuckerberg to his metaverse.
What kind of society are we building when those with greater economic and technological power plan their own escape instead of seeking collective solutions? If the creators of artificial intelligence are more afraid of artificial intelligence itself than we are, perhaps it's time to start asking more uncomfortable questions.
And demand clearer answers.
