Today we're going to talk about a topic that sounds a little crazy, but you really can't ignore it
The money in your wallet right now could completely disappear within 10 years.
I'm not saying this; it was said out loud by the richest person on Earth right now, Elon Musk, at a very important public event.
On November 19, 2025, in Washington, D.C., on the stage at the Kennedy Center for the U.S.-Saudi Investment Forum, Musk was sitting there, with Nvidia CEO Jensen Huang sitting beside him.
Two of the most influential people in today's tech world sat together, and Musk said this in front of the entire audience.
What did his original words mean?
He said that the monetary system you're familiar with will collapse within 10 to 20 years.
Then he gave a more specific milestone: the moment Tesla's Optimus robots begin mass production, which he expects to happen in 2028 or 2029.
He said that moment will be the first signal that the monetary system is starting to crack.
In other words, there are now less than three years left until the "signal appears" that he talked about.
Sounds like science fiction?
When I first saw this information, I also thought, fine, Musk is exaggerating again.
But wait, before you close this video, I want to first take everyone back and look at how accurate this person's past predictions have actually been
Because understanding that is the foundation of our entire discussion today.
If you find this content valuable, please give it a like and subscribe to the channel; that really matters to me.
Okay, let's sort through the predictions in Musk's history that were "mocked, but eventually came true."
The first one: reusable rockets.
In 2008, Musk said he would make a private company successfully launch a rocket for the first time, and not just launch it—he also said he would bring the rocket back and reuse it.
What was the industry's reaction back then?
Mockery.
Even engineers inside NASA said it was physically impossible. At the time, reusable rockets were widely regarded across the entire aerospace industry as something "impossible to do."
But in 2015, the video of a Falcon 9 first-stage booster landing vertically after launch was made public.
I still remember that video to this day; it went viral across the internet, and everyone was stunned.
And now? SpaceX has reused the same booster more than 20 times. This has already become common knowledge.
The second one: brain-computer interfaces.
In 2016, Musk announced that he wanted to implant chips into the human brain and connect the brain directly to computers.
The medical community's reaction at the time was extremely intense; many people believed that implanting foreign objects into brain tissue, even for experimental purposes, should not be allowed.
Musk's own goal was to achieve it by 2020.
2020 came, and nothing happened.
In 2021 and 2022, there was still no news.
People started saying again, see, he's bragging again.
Then in January 2024, Neuralink announced the real-life case of its first human patient.
A quadriplegic patient, lying in bed, moved a chess game on a screen using only his thoughts.
When you see that image, you get goosebumps.
It was indeed four full years later than the time he had set for himself, but in the end, it was still achieved.
The third one: satellite internet.
In 2015, he proposed launching thousands of satellites into low Earth orbit to provide internet coverage for the whole world.
At the time, mainstream telecom companies said: no company on Earth can afford this cost.
Now, the number of Starlink satellites has exceeded 9,000, with global coverage in more than 100 countries; even in the most remote mountain areas, you can still get online.
So do you see the pattern?
Musk's predictions often miss the timing, but he has never missed the direction.
His predictions only arrive late; they never fail to show up.
So this time, he says money will disappear, the job structure will be completely overturned, and the value system of assets will be rewritten
Okay, what is the core of Musk's prediction this time?
One word: robots.
More specifically, the humanoid robot Tesla is developing, Optimus.
Let's do a very simple but unsettling calculation.
The mass-production target price Musk set for Optimus is $20,000 to $30,000 per unit.
Converted to RMB, that's roughly 140,000 to 210,000 yuan.
Sounds expensive? Let's compare.
In the United States, hiring a regular warehouse worker—counting wages, social security, paid leave, health insurance, and management costs—the actual annual total employment cost is about $40,000 to $60,000, and that's a continuing yearly expense.
Optimus, by contrast, is a one-time investment, and then it can keep running indefinitely.
As for working hours, there's simply no comparison.
U.S. federal law says a worker's standard workweek is 40 hours, and anything beyond that requires overtime pay.
Optimus? 168 hours a week, around the clock, with no holidays, no lunch break, and no sick leave.
Just in terms of working hours, one Optimus can replace at least three to four human workers.
And that's not even the key point.
Musk says Optimus's production efficiency on specific tasks is more than five times that of humans.
Of course, that claim is based on mass-production costs falling below $30,000 and on operating in a standardized environment with close to 24-hour high utilization. He expects those conditions to be met between 2028 and 2029; more conservative analysts think it may not be until the mid-2030s.
But regardless of which time point it is, one logic remains unchanged
When the cost of introducing a robot is lower than the labor cost it saves, companies will replace humans on a large scale.
This is not up for discussion; it's the inevitable choice of the market.
22222 Many people hear that robots will replace humans and think that's something that will happen decades from now.
But I want to show you a case that is happening right now
Amazon.
In early 2026, The Guardian exclusively reported that Amazon planned to replace 500,000 jobs in the United States with robots by 2027 and reduce new hiring by at least 160,000 people.
Morgan Stanley analysts estimated that once this replacement is complete, Amazon could save up to $4 billion a year, which is roughly 29 billion yuan.
At present, Amazon has already deployed more than 1 million robots of various types across its global logistics warehouses.
One in particular is worth paying attention to, called Digit, the bipedal humanoid robot developed by Agility Robotics.
According to Amazon's publicly disclosed operating data, Digit's hourly operating cost, including electricity and allocated maintenance, is about $3.
The average hourly wage of U.S. logistics warehouse workers is between $18 and $22.
One costs $3, one costs $20—the same job costs only one-sixth as much as a human.
When company CEOs see that number, what decision do you think they will make?
An employee working at an Amazon warehouse in Georgia told The Guardian in an interview:
"The company is now making us take cross-training; people who used to do picking and loading are now being asked to learn robot maintenance skills. Because they know these jobs are going to be automated soon."
So you understand, right? This is no longer a prediction; it's a reality that is happening right now.
Of course, I also need to make one thing clear: robots cannot immediately replace humans everywhere.
They will first replace people in large factories and logistics centers with high utilization and simple, repetitive tasks, and then gradually spread to other industries.
For small and medium-sized businesses with changing job types and low utilization, robot maintenance costs and software update costs may exceed the labor costs they save.
So this replacement process happens in stages, but the trend is irreversible.
Okay, now let's work through what happens to the entire economic system once robots enter factories on a large scale.
First, production costs fall sharply.
The marginal cost of a factory product—that is, the additional cost of producing one more unit—will drop sharply because of robots.
What about output? It will grow explosively, because aside from adding more robots and expanding factories, there are almost no other bottlenecks.
Once output surges, the supply of goods in the market far exceeds demand, and prices begin to fall.
Clothing, furniture, food, cars, and electronics—all goods will trend downward in price.
This phenomenon has a name: deflation.
Over the past 50 years, the world's major economies have been fighting inflation.
Rising prices have always been one of the biggest headaches for central banks.
One of the Federal Reserve's core tasks for decades has been keeping inflation at around 2%.
But the problem now is that from the moment Optimus begins mass production, we may have to face the exact opposite situation: prices falling too fast.
Some people may say, aren't lower prices a good thing? Buying things becomes more affordable.
Wait—economically speaking, deflation is actually a very dangerous state.
Let me give you a simple example.
If you knew that something costing 100 yuan today would drop to 90 yuan tomorrow, would you buy it today or wait until tomorrow?
Of course you'd wait until tomorrow.
But if everyone is waiting, consumption stalls.
When consumption stalls, company sales fall, profits shrink, companies begin discounting to stimulate sales, prices fall further, and consumers keep waiting—this creates a vicious cycle that economists call a deflationary spiral.
Japan went through this in the 1990s and lost two full decades of economic growth.
So what should be done? Central banks' usual response is to cut interest rates, inject cheap money into the market, and stimulate consumption.
But if deflationary pressure continues, interest rates keep getting cut and eventually approach zero.
What does zero interest rate mean?
If you put money in the bank, you get almost no interest.
The concept of "saving" will lose most of its meaning.
When Musk said in an interview, "You don't need to save for retirement," he was based on this logic.
Okay, speaking of this, let's talk about the most practical question
Is your job safe in this wave?
The World Economic Forum released a Future of Jobs Report in January 2025, and the data is very interesting.
The report says that by 2030, there will be 170 million new jobs created globally, but at the same time 92 million jobs will disappear.
That means the number of new jobs exceeds the number of lost jobs by about 80 million, which sounds like good news?
But there's a very critical detail here
The jobs that disappear and the jobs that are created are aimed at completely different groups of people.
A customer service representative in a call center is replaced by AI; she will not automatically become an AI engineer.
There is a huge skills gap between those two things.
So which occupations will disappear first?
Microsoft CEO Satya Nadella once publicly said that 30% of his company's code is now written by AI.
Compared with the 2022 peak, hiring of new developers aged 22 to 25 in the United States has fallen by a full 20%.
In less than five years, one out of five people has quietly disappeared.
Dario Amodei, the founder of Anthropic, the company that develops Claude, said something even more direct on a podcast:
"Within five years, half of the entry-level white-collar jobs may disappear."
"Half" means the very concept of your first job may disappear altogether.
Professor Kim Dae-shik of the Korea Advanced Institute of Science and Technology has repeatedly emphasized a point in multiple public talks:
The first people AI takes away are neither the worst nor the best, but the "average performers."
Outstanding people will use AI as a tool to make themselves stronger.
Average repetitive work will be directly replaced by AI.
The middle layer will disappear.
The occupations with the fastest decline in the World Economic Forum report include: cashiers, ticket clerks, simple administrative assistants, and data entry clerks.
These are all highly repetitive roles that require standardized judgment.
What about lawyers? They're no exception.
Writing complaints, searching case law, reviewing contracts—AI can already do the work of finding and organizing legal knowledge better than humans, more accurately than humans, and at very low cost.
What about doctors? Multiple medical studies show that work such as image diagnosis, pathology analysis, and assistive diagnosis is highly likely to be replaced by AI.
33333 Google's top futurist, Thomas Frey, has predicted in multiple talks: by 2030, 2 billion jobs will disappear worldwide.
But he also said something that left a deep impression on me:
"As long as problems exist, the jobs that solve problems will always remain."
He also said: "By 2030, the biggest company on the internet will be an education company."
Because in this era, the ability to learn quickly and adapt to new environments will be worth more than any specific vocational skill.
Okay, just saying which jobs will disappear is not enough; let's talk about which jobs will stay and even become more valuable.
In the World Economic Forum report, the occupation with the largest growth in absolute numbers, ranked first, what do you think it is?
It's agricultural workers. Driven by the climate crisis and the green transition, demand for food security and sustainable agriculture is surging.
Next are nursing, construction, and education.
What these occupations have in common is that they require physical presence on site, or they are centered on emotional trust between people.
The fastest-growing occupations are big data specialists, AI and machine learning engineers, fintech specialists, cybersecurity experts, and so on.
But I think more important than the specific job list is understanding the two core axes that futurists are collectively pointing to.
The first axis: people who can use AI.
At the 2021 University Innovation Forum, Thomas Frey emphasized that future education must use AI to greatly increase learning speed.
This is already happening—one person plus AI tools can produce results that used to require a team of dozens.
People like that will not disappear; they will only become more and more valuable.
They are the group whose "1x productivity becomes 10x productivity."
The second axis: people who do things AI cannot do.
Professor Kim Dae-shik emphasized a very profound point: if AI becomes too perfect, humans lose the chance to fail. And without failure, you cannot accumulate true judgment.
The experiential judgment, empathy, ability to build trust, and ethical decision-making developed through failure—AI can simulate these, but it cannot truly feel them, and it cannot bear responsibility.
Parents caring for a child for the first time, doctors facing complex patient emotions, leaders who truly mobilize a team, people who mediate community conflicts—these roles operate through relationships and trust, not data and algorithms.
As Thomas Frey said, someone entering the workforce in 2030 may change careers 8 to 10 times in a lifetime.
Rather than clinging to one occupation, the ability to learn quickly and adapt to new environments will become the core competitiveness of this era.
Okay, let's return to the core question.
If money really becomes ineffective, what will become the new carrier of value?
Musk's answer is energy.
At first glance, that's very abstract—how do you trade things with electricity?
But let me explain Musk's logic.
What is the biggest structural problem with paper money right now?
It can be increased infinitely by a political decision.
The Federal Reserve can hold a meeting and decide to print another trillion dollars.
That's why currency keeps depreciating, and that's why inflation exists—because the amount of money issued is not constrained by physical resources.
But energy is different.
You can't "issue" electricity in a conference room.
You have to install solar panels, build nuclear reactors, and invest real material resources to produce electricity.
That gives energy itself a kind of "anchoring" function.
The reason Bitcoin is recognized by many people follows the same logic: mining one Bitcoin consumes a large amount of real electricity and computing power. That is completely different from fiat currency logic, where you can create money just by writing a number.
Of course, energy is not the only answer; futurists also point to several other types of valuable assets.
In his book The Future Is Faster Than You Think, Peter Diamandis wrote that when the exponential development of multiple technologies begins to converge, it will not only overturn existing products and services, but also destroy the entire social structure that supports those products and services.
He believes future wealth will concentrate in the hands of those who first see the changes clearly and position themselves early.
Ray Kurzweil, from another angle, points out that technological development is not linear, but follows an exponential growth pattern. In this trend, the truly valuable assets are: knowledge itself, the systems that carry knowledge, and the energy that drives those systems.
Taking these futurists' views together, let me sort out the truly valuable asset structure of the next era for everyone:
First category: energy production capacity.
Solar power, small modular nuclear reactors, battery storage systems. The ability to produce electricity independently will become the foundation of national competitiveness and personal assets. This is not optional; it is a core infrastructure asset.
Second category: data, especially high-quality data with emotion and narrative.
This is something many people overlook. Training AI models requires huge amounts of high-quality content data, especially content with human emotion, cultural storytelling, and professional depth. South Korean film and television content and K-POP-related data are already among the most expensive categories in the AI training data market; that's no accident.
Third category: ownership of the AI and robot systems themselves.
Over the past hundred years, the wealth carriers of the industrial age were land and factories. In the next era, the wealth carriers will be AI systems and energy infrastructure. The people who truly get the dividends of this era are those who see the changes first and take position early.
Fourth category: trust and relationships between people.
The more technology advances, the scarcer direct trust between people becomes, and the more valuable it becomes. This is not soft motivational talk; it is a quantifiable change in value.
Okay, speaking of this, I also have to tell everyone that the reason Musk is talking about all this so openly and so loudly is tied to a very concrete real-world issue.
That is the U.S. national debt problem.
As of 2025, U.S. national debt is about $38.4 trillion.
What does that mean?
444444 This figure is more than 80 times South Korea's entire annual budget, more than twice China's GDP, and about 1.3 times the U.S.'s own GDP.
Under normal circumstances, this money simply cannot be paid back.
So what solution does Musk propose?
The strategy he proposes is called **"debt dilution"**—not actually repaying the debt, but making the debt seem smaller and smaller relative to the overall size of the economy.
Let me give you an analogy you can understand.
A small restaurant with annual revenue of 1 million yuan owes 500,000 yuan; the pressure is huge, and it could go under at any time.
But if that restaurant expands into a chain and becomes a large group with annual revenue of 10 billion yuan, then the 500,000 yuan debt would account for only 0.0005% of total revenue—basically imperceptible.
The future of the United States that Musk describes follows this logic.
Through AI and robots, make U.S. GDP grow explosively and raise productivity to a new order of magnitude.
At the same time, robots massively replace human labor, goods prices fall, deflation arrives, and interest rates are forced to zero.
Once interest rates go to zero, what is the most direct benefit?
The U.S. currently pays more than $1 trillion a year in debt interest alone.
At current exchange rates, that is more than 7 trillion yuan per year.
If interest rates go to zero, that annual interest expense of more than $1 trillion will shrink significantly.
This is Musk's full script:
Use AI and robots to boost GDP, use deflation to push interest rates down, let debt shrink relative to the economy, and eliminate the interest burden at the same time—the United States solves its debt crisis in one fell swoop and firmly grasps technological hegemony in the new era.
Sounds too perfect, doesn't it?
Of course, whether this script can be fully implemented according to plan still depends on many variables.
But this direction is already moving forward.
While talking about all this, Musk himself also raised two warnings that should make people alert.
The first warning is the risk he calls "pet humans."
When AI surpasses humans at every level, people will increasingly get used to letting AI decide what to eat, whom to meet, and what decisions to make.
Over time, humans' autonomous judgment ability will begin to deteriorate.
This is consistent with Professor Kim Dae-shik's view: if AI is too perfect, humans lose the chance to fail and grow.
Without experiencing failure, there is no real judgment.
This is not alarmism; it is a cognitive decline risk that is already happening.
The second warning is the population problem in South Korea that he specifically singled out.
Musk cited data saying that South Korea's birth rate is extremely low, and if the current trend continues, three generations from now South Korea's population may be only 3% to 4% of what it is now.
He said this is one of the threats to all humanity, because a highly developed, technology-intensive population is disappearing.
But here's a paradox I find very interesting.
In the new economic structure Musk describes, South Korea is actually in a very unique position of advantage.
Why do I say that?
Because South Korea is almost the only country in the world that simultaneously holds the three core assets of the AI era.
First, high-bandwidth memory, HBM, High Bandwidth Memory.
SK Hynix and Samsung Electronics together account for more than 90% of the global HBM market.
Without South Korea's HBM, Nvidia's most powerful AI chips cannot fully realize their performance.
This is a strategic resource with no substitute.
Second, nuclear power design and construction capability.
The energy consumption of AI data centers is astonishing.
According to data from the International Energy Agency, by 2026 global data center electricity consumption is expected to exceed 1,000 terawatt-hours, equivalent to Japan's total annual electricity use.
And South Korea is currently the only country in the world that can complete the construction of a large nuclear power plant on schedule and within budget—there is a real-world case: the Barakah Nuclear Power Plant built in the United Arab Emirates.
Third, content data with emotion and narrative, that is, South Korea's cultural content—K-dramas, K-movies, K-POP.
This type of content is the most expensive in the AI training data market because it contains rich human emotion, narrative structure, and cultural context, making it the highest-quality material for AI models to learn human language and emotional expression.
So you see, population decline is South Korea's obvious weakness, but in the core position of the next economic structure, South Korea actually holds several extremely scarce cards.
Okay, after talking so much about trends and predictions, we still have to return to the most practical question
As ordinary people, what should we do in the face of this change?
Taking the views of major futurists together, I've organized four core directions for everyone.
The first direction: master the ability to use AI.
This does not mean you need to learn programming; it does not mean you need to become an engineer.
It means you need to learn to use AI tools as your own "super assistant" so that it can amplify your productivity.
This is already a reality—an individual developer who knows how to use AI can make products that used to require a team of dozens.
Marketers who know how to use AI can produce work at 5 to 10 times the efficiency of ordinary colleagues.
The ability to learn new things quickly will become the core competitiveness of this era.
The second direction: shift from "selling time" to "owning systems."
Trading time for money is the logic of the industrial age.
In the AI era, what is truly valuable is: owning systems that use AI, positioning early in energy-related assets, and producing and owning data yourself.
These are assets that can "make money while you sleep," not time-for-money exchanges that stop generating income once you stop working.
The third direction: strengthen core capabilities that AI cannot replace.
That is the two axes I mentioned earlier
Leadership that gets people to act.
The ability to build trust and empathy.
The ability to make judgments in complex, ambiguous, emotionally charged situations.
These skills will only become more valuable in the AI era, because they are becoming more and more scarce.
The fourth direction: allocate core assets.
555555 I want to make a special statement here: this is not investment advice; everyone's situation is different, and investing involves risk.
But from a trend perspective, the asset directions that futurists repeatedly point to include:
AI core companies, such as Nvidia, Alphabet, and Microsoft;
next-generation energy companies, such as companies related to small modular nuclear reactors, battery storage firms, and solar companies;
robot-related companies, including Tesla's Optimus and the robot component supply chain;
infrastructure companies, including data center operators, power grid companies, and satellite communications companies.
In the industrial age, the foundation of wealth was land and factories.
Okay, let me share my personal judgment.
The specific milestones Musk gives may not be exact. Mass production of Optimus may slip from 2028 to 2032, and changes to the monetary system may not be a "collapse" but rather a long, gradual process of becoming ineffective.
But the direction has already begun to emerge.
You can see that hiring of young developers in the United States has fallen by 20%; that's already reality.
Amazon's plan to replace 500,000 jobs with robots is already reality.
AI writing code, AI drafting legal documents, AI diagnosing images—these are all already reality.
I think Thomas Frey's line is the best summary of the whole discussion:
"As long as problems exist, the professions that solve problems will always remain."
The ultimate winners of this era are the people who always stand on the side of "solving problems," not the people waiting to be replaced.
Okay, that's all for today's content.
If you think this episode is valuable to you, please give it a like; that will help more people see this content.
If you think there are friends around you who should know this information, forward it to them—that's one of the most valuable things you can do for them.
See you in the comments
I'm really curious: after watching this episode, do you think your job is safe? And how are you planning to respond to this change?
Leave a comment; I'll read every reply carefully.
See you next time.
