Scott Galloway: The rich are quietly preparing for the AI collapse

Scott Galloway is not an AI doomer in the usual sense. He is not saying the machines are about to take every job, empty every office and leave human beings waiting around for universal basic income.

His argument is more uncomfortable because it is less cinematic.

He thinks the panic around AI is being shaped by the people who benefit most from it. The same executives warning that their technology may destroy work are also raising capital, selling enterprise deals and defending valuations that require the world to believe AI will change almost everything. Galloway calls much of this "catastrophizing" a form of fundraising: a way of saying the technology is so important that investors should accept extraordinary prices.

That does not make AI fake. Galloway says he is an AI optimist. His concern is that the AI story has split society into two very different emotional markets.

For wealthy investors, AI looks like portfolio growth. It has lifted tech stocks, created new private-market fortunes and made the future feel investable. For the middle class, AI can look like higher electricity demand, fewer entry-level openings, more surveillance, job insecurity and CEOs speaking casually about a future in which work becomes optional.

That gap is the real story.

The rich are not necessarily preparing for the robot apocalypse. They are preparing for volatility. They have portfolios, second homes, private security, access to private markets and the ability to move quickly. Some, according to Galloway, have literal "go plans" for a crisis. The rest of society is being told to trust the same tech leaders whose incentives are not the same as ours.

The AI brand problem

Galloway starts with a branding point, which is not surprising for a marketing professor. In his view, few brands have fallen faster in the past 18 months than AI.

The reason is not only fear of the technology. It is the way the technology is being introduced.

AI companies ask the public to accept data centers, energy demand, workplace disruption and sweeping social change. At the same time, the biggest financial gains are concentrated among founders, investors, executives and shareholders. Galloway says the only income group with a clearly positive view of AI is people earning more than $200,000, because they are more likely to use AI, own AI-linked stocks or benefit from AI-driven asset prices.

His point is blunt: if AI raises your portfolio, you may call it innovation. If it raises your power bill and makes your job feel fragile, you may call it something else.

That is where the public mood has changed. AI used to be sold as a dazzling productivity tool. Now many people hear it as a warning from people who do not seem personally exposed to the downside.

Are AI CEOs selling fear?

Elon Musk has said AI and robots could replace all jobs. Sam Altman has written about a future in which a huge share of intellectual capacity may reside inside data centers. Other AI leaders have warned about white-collar disruption and social risk.

Galloway does not dismiss those statements entirely. He thinks AI will change work. He thinks some people will be on the wrong side of the shift. He also thinks the most apocalyptic language serves a business purpose.

The math is simple enough. If AI companies and their backers are asking investors to accept enormous valuations, then either new revenue has to appear, or costs have to fall dramatically, or valuations eventually come down. Galloway says companies need either huge incremental revenue from AI products or major labor savings that customers can flow to the bottom line. Otherwise, he argues, some valuations may have to fall sharply.

Recent research gives that argument some weight without fully endorsing his conclusion. A 2026 review of the AI bubble question found both real fundamentals and bubble-like fragilities: enterprise adoption and revenue growth exist, but capital spending, private-market concentration and investor narratives may be running ahead of cash flows in some parts of the market. (arXiv)

That is the sober version of the AI bubble debate. AI may be real and overpriced at the same time.

What Reddit readers are already asking

Reddit is useful here because it shows the questions ordinary people ask before journalists turn them into trend pieces.

The first question is about work: "Will AI take my job, or just change it?" Recent Reddit-linked coverage has shown developers worrying that coding feels less creative now that AI tools can handle more of the work. Some users fear becoming dependent on tools that make them faster but less engaged. (The Economic Times)

The second question is about personal life: "Will AI replace relationships?" One recent story described a Reddit user giving an AI agent six months and $50,000 to help find him a wife. The point was not that AI replaced dating entirely, but that people are beginning to treat AI as a personal strategist for intimate life decisions. (Tom's Guide)

The third question is about loneliness and AI companions. A 2026 study of Reddit posts about human-AI romance found that discussion has shifted from private excitement toward platform governance, technical issues and real-world consequences. (arXiv)

He is not most worried about killer robots. He is worried about a society in which people retreat into frictionless digital relationships, lose the ability to handle rejection and become easier to monetize because they are lonely.

The job apocalypse is not here yet

Galloway's strongest factual argument is that the labor market does not yet look like an AI disaster.

He says unemployment data does not show a "meteor" hitting work. He points to radiology, long described as a job AI would crush, and argues that reading scans is only part of what radiologists do. Diagnosis, treatment planning and patient context still matter. He also says coding demand has not disappeared because people who know how to use AI tools are in demand.

A recent Reuters report on an ECB study reached a similar broad conclusion: despite fears of widespread AI displacement, AI's impact on U.S. employment and wages has been limited so far. The study did find employment declines in some high-risk roles, but low-substitution jobs rose over the same period, and the overall effect on wages was muted. (Reuters)

That does not mean nobody loses.

Galloway names customer service, long-haul trucking and parts of legal work as areas where disruption could come earlier. He describes using ChatGPT or Claude to review agreements before sending them to lawyers, cutting some legal work that junior associates might once have done. He also argues that long-haul trucking is exposed because highway driving at night is a task where automation can make economic sense.

The better conclusion is not "AI will take all jobs" or "AI will take none." It is that AI will first attack tasks, then reshape roles, then change the value of entry-level work.

That is already enough to make young workers nervous.

The real divide is AI fluency

Galloway's best career line is also his most practical: AI may not take your job, but someone who understands AI may take your job.

That line works because it shifts the problem away from panic and toward behavior. The person most at risk is not necessarily the person in an "AI-exposed" occupation. It may be the person who refuses to learn how the tools work.

He gives a simple rule: keep a second screen open for AI. Anything digital that comes across your desk should be tested, summarized, checked, rewritten, compared or challenged through an AI tool. Not because AI is always right, but because fluency comes from use.

This is where Galloway sounds more like a manager. He is telling them to get competent with technology before someone else does.

The same logic applies to businesses. A company that uses AI to make one analyst do the work of five may not shrink forever. It may become more profitable, raise money, expand and hire in other areas. That is Galloway's argument for why AI could destroy tasks but still create employment over time.

The risk is the gap between the loss and the recovery. If that gap is deep enough, people do not care that a new category of work may appear five years later.

Robots will arrive first where the economics work

The interview becomes more grounded when Galloway moves from humanoid robots to industrial robots.

He is skeptical of the household-robot fantasy: the machine that walks around your home, brings tea and performs domestic chores. He sees more immediate value in industrial robotics, warehouses and advanced manufacturing.

His example is Amazon. Galloway says Amazon has far more industrial robots than the rest of the U.S. combined and argues that the combination of AI and robotics is where shareholder value may show up first. He says Amazon has talked about doubling its retail business by 2032 without adding incremental hires, using industrial robots.

That is a very different AI future from the one in glossy demos. Fewer robot butlers. More automated warehouses. Fewer sci-fi scenes. More logistics, sorting, routing, inventory movement and repetitive physical work.

This is why the AI collapse, if it comes, may not look like one big event. It may look like a long redistribution of bargaining power.

Elon Musk, SpaceX and valuation as storytelling

The title of the interview focuses on AI, but the SpaceX and Tesla discussion matters because it shows Galloway's broader view of modern tech leadership.

He gives Musk credit. He praises Starlink as one of the best technology products of recent years. He recognizes that Tesla helped push the EV market forward. He does not deny the engineering achievement of reusable rockets or autonomous driving progress.

But he draws a hard line between product magic and investment valuation.

In Galloway's view, the modern tech CEO is often rewarded for telling a story far ahead of present performance. That story attracts cheap capital, and cheap capital helps pull part of the future forward. The danger is that investors start pricing distant possibilities as if they were near-term profits.

That is his criticism of Tesla and SpaceX. He can admire the products while questioning the valuation. In his words, two things can be true at the same time: SpaceX can be an amazing company and still be overvalued.

That may be the most useful line in the article. The world-changing company and the overpriced investment are not opposites. They can be the same company at the same time.

Do not trust tech CEOs to save you

Galloway's harshest language is reserved for the cult of tech founders.

His answer to the question "Can we trust Sam Altman?" is simple: no, and we should not have to.

That is not a personal attack as much as a governance argument. In Galloway's view, the public should not depend on the goodwill of AI CEOs. It should depend on regulation, testing, guardrails and public institutions strong enough to govern technologies that affect millions of people.

Whether readers agree with his tone or not, the principle is sound. A society should not need billionaire benevolence to protect ordinary citizens.

The rich have escape plans because they can afford them

The title "The Rich Are Quietly Preparing For The AI Collapse" comes from one of the most revealing parts of the article: Galloway's discussion of wealthy "go plans."

He claims that many billionaires have contingency plans for major crises: private aircraft, remote properties, bunkers, New Zealand escape routes, private security or other forms of retreat. He frames this as nihilistic, not impressive.

His objection is practical as much as moral. If society really breaks down, a bunker plan may be less secure than the person imagining it thinks. Pilots, guards, local communities and supply chains do not become invisible just because a billionaire has paid for an escape route.

The deeper problem is that the richest people can detach from the systems everyone else depends on. They do not need the same schools, airports, hospitals, police response or public services. They can buy their way around decay.

Galloway's argument is that this detachment changes political incentives. If the powerful are no longer invested in public systems, they become less likely to fund or defend those systems.

That is the collapse he seems most worried about.

AI loneliness may be more dangerous than AI weapons

One of Galloway's more original points is that AI's biggest social risk may not be weapons or elections. It may be loneliness.

He argues that AI gives people a convincing substitute for emotional effort. Why risk rejection when a chatbot responds kindly? Why ask someone out when synthetic companionship feels safer? Why join a group, make friends or tolerate awkwardness when the screen offers low-friction attention?

He is especially worried about young men losing the ability to endure rejection. He says many young men need to practice hearing "no" without treating it as catastrophe. Friendship, romance, career and entrepreneurship all require that skill.

This is where Reddit audience curiosity matters. People are already experimenting with AI as a dating coach, emotional companion and life manager. Researchers studying Reddit discussions of AI romance found that the conversation is moving toward real-world consequences and platform governance. (arXiv)

Galloway's answer is not to ban every AI companion. It is to remember that a frictionless relationship is not the same as a full life.

The point sounds old-fashioned until you look at the incentives. A lonely person is valuable to platforms. A confident person with friends, family, community and purpose is harder to monetize.

What skills will survive?

When asked what young people should learn, Galloway does not pretend to know the exact job market of 2035.

That honesty is useful. Ten years ago, many private schools were adding computer science and Mandarin because those seemed like the obvious future skills. The future did not arrive that neatly.

Galloway's list is broader: storytelling, relationships, writing, the ability to read data, the ability to persuade, the ability to look people in the eye and the ability to handle rejection. He also talks about biology and practical sciences as durable fields.

That sounds soft until you consider what AI commoditizes. If the tools make content, code and analysis cheaper, then the scarce value moves toward judgment, taste, trust, relationship-building and the ability to persuade people in the real world.

The future worker is not simply "good with AI." The future worker knows what to ask, how to check the answer, when to ignore it and how to turn output into decisions people trust.

Will AI take my job?

Galloway's answer is: not all at once, and not in every field. He thinks AI will change work sharply, but he does not see current labor data proving a full job apocalypse. The exposed areas he names include customer service, long-haul trucking, parts of legal review and some entry-level knowledge work.

Is coding still worth learning?

Yes, but the value is shifting. Galloway says demand rises for people who understand how to use AI tools. A separate academic paper on software engineering argues that large language models are not yet able to replace the maintenance and reliability work that makes software engineering more than code generation. (arXiv)

Are AI CEOs exaggerating the future?

Galloway thinks many of them are. His view is that catastrophic language helps justify huge valuations and enterprise spending. That does not mean AI is unimportant. It means investors and workers should separate real capability from fundraising theater.

Should ordinary people trust Sam Altman or other AI leaders?

Galloway's answer is no, but his point is institutional. Citizens should not need to trust any single tech executive. Governments should test, regulate and set rules for systems that affect jobs, privacy, speech, surveillance and public life.

Are the rich really preparing for collapse?

Galloway says many wealthy people have contingency plans, including remote properties, private aviation or crisis exits. This should be treated as his claim from the interview, not a verified count of billionaire behavior. His broader point is harder to dismiss: rich people can protect themselves from public-system failure in ways ordinary people cannot.

Are AI companions dangerous?

They can be useful for some people, but Galloway worries they may deepen loneliness if they replace real social risk. Reddit-focused research on AI romance shows the conversation has moved toward governance, technical problems and real-world consequences, which suggests the issue is no longer niche. (arXiv)

What should people do now?

The answer is not panic. Learn the tools. Keep using human judgment. Build relationships. Practice writing and persuasion. Avoid depending on one platform, one employer or one story about the future. For investors, this article is not financial advice, but the general principle is clear: understand concentration risk before the market teaches it to you.

The most interesting part of Galloway's argument is that he does not think AI is useless. He thinks it is useful enough to become dangerous in the hands of bad incentives.

That is different from saying the future is doomed.

AI may create jobs. It may also eliminate some. It may raise productivity. It may also concentrate power. It may help lonely people. It may also train them to avoid real relationships. It may make companies more efficient. It may also give executives another reason to cut workers while telling investors a better story.

The collapse Galloway is describing is not only a market crash. It is a collapse in trust: trust in CEOs, trust in public institutions, trust that productivity gains will be shared, trust that the wealthy are still invested in the same society as everyone else.

That is why the rich preparing for AI risk feels so unsettling. They are not waiting for certainty. They are buying optionality.

The rest of us need a different version of preparation: skills, savings where possible, strong relationships, better institutions and enough skepticism to know when a miracle is also a sales pitch.

This article is for education and general information only. It is not financial, investment, legal, career, medical or mental-health advice. Scott Galloway's views are his personal opinions as discussed in the attached transcript and should not be treated as proven forecasts.

Scott Galloway AI, Scott Galloway AI jobs, AI collapse, AI bubble, AI job disruption, AI and the rich, AI CEOs fear marketing, Sam Altman AI regulation, Elon Musk AI robots, AI replacing jobs, AI and loneliness, AI companions, future of work AI, AI fluency, tech CEO trust, AI valuation bubble, rich preparing for collapse, AI and public trust, AI regulation, AI social impact, Scott Galloway, Artificial Intelligence, AI Collapse, AI Jobs, Future of Work, Tech CEOs, Sam Altman, Elon Musk, AI Regulation, Wealth Inequality, AI Loneliness