The Fed Chair Just Admitted The Jobs Aren't Coming Back — Here's What Happens To Your Career Next

AI has already changed the world in a way that most people simply do not understand yet. Last month, the chair of the Federal Reserve walked up to a podium and admitted what most people could already feel. Effectively, there's zero net job creation in the private sector. That's a direct quote. He didn't say growth slowed, he didn't say we had a soft patch, he said we had zero growth. December jobs have been revised to a loss, the gains the public were told we had in January were actually off by 69,000. In February, the economy shed 92,000 jobs, and when you add it all up, 2025 was the weakest year for job growth outside of a recession since 2003. Now, there are two possible explanations for what's happening. The first is this is just a natural cycle. There's a ton of uncertainty right now in the global economy, and the street rewards companies that are fiscally disciplined, so companies cut. Now, if that's true, the jobs will come back as they always do. But the second explanation is that something structural has changed. People are being replaced by AI that is getting cheaper, faster, and better by the minute. By the end of this video, you're going to know exactly which one of those is based on the actual data. I'm going to show you what it means for your job, your income, and your options, and exactly what you can do about it. I'm going to prove all of that in four parts. Part three is my most dire warning, so do not skip that. Welcome to part one. You're being force-fed narratives. Plan based on the data. Marc Andreessen co-invented the web browser. He manages today over 90 billion dollars in assets at one of the most influential venture firms on Earth. And recently, he went on record saying that all of the supposed AI-driven layoffs are, and I quote, >> This entire labor displacement thing is 100% incorrect. It's completely wrong. Now, he's got a point. But he also has an incentive, or at least a blind spot. Here's his argument. Between 2020 and 2022, companies went on a hiring binge. It's very true. COVID forced everyone to work remotely, but remote work kills discipline and productivity, but money was essentially free, given that interest rates were near zero, so companies just threw headcount at the problem. It actually became a kind of vanity metric, how many employees you could rack up. Tech companies, in particular, ballooned their workforces in ways that made no fundamental business sense. Then, interest rates went from zero to 5% in just 3 years. It was the fastest rate increase in modern history. Suddenly, every large company had to completely re-plan their finances, and the cost of carrying all those extra bodies became impossible to justify. But laying people off because you over-hired is terrible optics. And if you're a public company, it could be bad for the share price. But if you can cut the staff and say it's because you're using AI, the street loves it. They assume you're just being disciplined while also planning for future growth. And thus, many companies, using this playbook, saw their shares go up in value. Andreessen's argument is that the layoffs you're seeing that are being blamed on AI are actually just companies finally addressing some previous bad decisions from COVID, finally right-sizing their company, and then using the AI excuse is good PR. He estimates that the majority of large companies are still overstaffed by 25%. Some are overstaffed by 50%, and a handful are overstaffed by 75%. And he believes that AI is just the silver bullet excuse, as he calls it, the clean narrative that lets executives avoid admitting they hired recklessly during COVID. It's a compelling argument if you just glance at the data, but not if you look at it deeply. New York became the first state to require companies to disclose whether AI contributed to their layoffs. They added it to the WARN filings that they require companies to file when they're about to lay people off. Of the 162 companies that filed notices, the overwhelming majority of them gave no indication that AI was the reason. But many did cite the general economy. And that's why I think Andreessen is right about many companies right now making noise to the public saying that it's about AI, but when they go talk to regulators, they actually leave AI out of the discussion altogether. They want investors to believe that they're being disciplined, but when they're talking about legal documents and regulators, all of a sudden, it's the economy and market conditions. But pointing out that companies would do that is far different than saying AI simply isn't impacting the job market now and into the future, which it most certainly is. Because when you dig into the question of whether or not AI is impacting the job market, the data starts telling a very clear story. It is completely transforming the job market already, and AI is just getting warmed up. Here's what we know already. A team at Stanford tracked employment in AI-exposed occupations from the moment ChatGPT launched in November of 2022 through early 2026. What they found was a 16% relative decline in employment for workers between 22 and 25 years old in the most AI-exposed roles. Entry-level software developers are down roughly 20%. Call center workers are down 15%. Mid-career workers, they were flat. Now, this kind of loss in the entry-level, flat at the middle job restructuring is the kind of thing that nobody holds a press conference about. It doesn't move a stock price. These are just positions that get identified as low-productivity jobs, and they stop getting filled. No big deal for the company, but for people trying to get into the workforce or trying to work their way up, it's devastating. And that is exactly the pattern that you'd expect to see if AI is being used as a tool by the most intelligent, most productive workers in an organization. If AI were truly autonomous, you'd actually see a different pattern. But it's not autonomous yet. As of today, AI is best thought of as an amplifier. But you only give an amplifier to people who are already performing well. And if that's true, then we'd expect the data to show that senior-level workers have more employment opportunities now than before AI. And lo and behold, that's exactly what the data shows. Just at the very moment that entry-level jobs are beginning to disappear and mid-level jobs are growing stagnant, senior-level positions are growing. Give an amplifier to a senior-level employee, and the productivity increases that you're likely to get mean you need fewer people to get more done. AI is just absolutely gobbling up those entry-level and mid-level jobs like a paralegal. The bottom rung of the cognitive labor market, the people whose job consists mostly of grunt execution of a task that was figured out and systematized by somebody else, but still needs to be executed on, that's just started disappearing wholesale from the workforce. But that's exactly how people get started, and they start moving, and it's just going away. This stark reality is starting to give us a glimpse into what's going to happen to the labor market. So, whether Andreessen is right or not about some of the more high-profile layoffs, the structural change that AI is already bringing to the workforce is visible right there in the data for anyone to see. Even if AI ends up creating more jobs than it destroys, which I don't actually believe, but that's a question of time scale, for now, even if we assume that it's true and AI is going to create more jobs, the jobs that it creates will be on a totally different kind of career ladder. The world in the near future simply isn't going to look anything like it does now. Welcome to part two. The first billion-dollar, one-person company is already being built. In 2024, the CEOs of the world's most powerful AI companies created a betting pool. This is a real story. Their wager, that AI was becoming so powerful that it was only a matter of time before it allowed a single person to build a billion-dollar company. The bet wasn't a question of if it would happen, but only when. They may have just gotten their answer. In September of 2024, a guy named Matthew Gallagher launched a telehealth company called Med Vi out of his living room with no employees and only $20,000 in startup capital, far less than most young people have just in student loans. There was no venture capital involved. He had a laptop and roughly a dozen AI tools. In 2025, his first full year in business, Gallagher and his AI team generated $401 million in revenue at a 16.2% net profit margin. That would be extremely rare for any company. So, the fact that it was achieved by one guy and his AI crew is absolutely insane. And given how quickly he amassed that revenue and how profitable the company is, odds are, even in year one, when Gallagher was all by himself, the company would have been valued at over a billion dollars, closing the betting pool just 1 year after it began. His closest competitor in the space, Hims & Hers, does similar work, but with 2,442 employees, and runs on a 5.5% margin. Gallagher is running nearly three times that margin even after hiring his first employee in year two, his brother. You You prior to AI, for all of human history, every major technological revolution followed the same pattern. It killed the old jobs, but it created new ones. The loom wiped out hand weavers, but employed a new generation of factory workers doing way more. The steam engine killed entire trades, but gave birth to entirely new industries. Electricity, the assembly line, the internet. Every single one of them destroyed a massive amount of jobs in one category, only more in another. And every time economists warned about mass unemployment. Every time though, they were wrong. New categories of work emerged that nobody could have predicted, and more ultimately ended up employed than before. That track record is so consistent that economists gave the fear of a new technology a name. They call it the lump of labor fallacy. The mistaken belief that there's a fixed amount of work in the economy, and that if a machine takes some of it, there's less left for humans. It's been wrong every time it's been applied without a single exception. So, here's the uncomfortable question. What if this time is actually different? Not because the technology is bigger, but because it's categorically different from everything that came before it. Every previous technology replaced physical or mental labor. The loom replaced hands. The tractor replaced backs. Even computers were dumb machines that merely gave new tools to human minds, but they certainly didn't replace actual intelligence. That is until AI. AI is the first tool we've ever created that can think and execute. So, if you think of AI as just another labor-saving device, I get why you would think that it's going to be like every technological revolution that came before it. But that would be what's known as a categorical error. AI is a zero-to-one moment. Non-human strength and computation has existed for a long time, but until now, human-grade intelligence was strictly the domain of well, humans. But not anymore. AI doesn't just do a task, it replaces the intelligence above the task. The pattern recognition layer. The reasoning that figures out what needs to be done in the first place. And that changes the game. Now, I know better than to say that this time is different. So, to the lump labor fallacy, I'll say this. I'll take it on faith that in the long run, it will create more jobs than it destroys. But when you factor in that this is an entirely new category of tool, just as the world was fundamentally altered by the Industrial Revolution, the internet, and modern finance, the world will be unrecognizable on the other side of AI. We'll get right back to the show, but first, let's talk about the work you're doing that you shouldn't be. You just spent 20 minutes logging call notes, then you spent another 10 figuring out what to do next. That's 30 minutes of manual work AI could be handling for you. Multiply that across every call your team takes, that's hours and hours of wasted time every day. Aircall is the number one rated business phone system on G2 with over 3,000 reviews, because it's more than just a phone. It's built for how modern teams actually work. More than 90,000 businesses rely on Aircall because it's a smart system, not just a phone system. Your entire team shares one business number, calls, texts, voicemails, transcripts, and contact details all live in one clean view. With full context at your fingertips, your team communicates faster, stays aligned, and delivers a more personal experience. Try Aircall for free, plus 20% off your first 6 months at aircall.com/impact. No missed calls, no missed customers. And now, let's get back to the show. We will be right back to the show, but right now, I want to talk about the AI mistake most businesses are making. Everyone's racing to adopt AI, rightfully so, but AI is only as good as the data behind it. And if your data lives in five different systems that don't even talk to each other, your AI is working on a broken foundation. NetSuite by Oracle is the number one AI cloud ERP trusted by over 43,000 businesses. It brings your financials, inventory, commerce, HR, and CRM into one unified system. One source of truth. Now, with NetSuite AI connector, use the AI of your choice to tap into your actual business data. This isn't a bolted-on tool, it's AI built into the system that runs your business. If your revenues are at least in the seven figures, get NetSuite's free business guide demystifying AI at netsuite.com/theory. The guide is free at netsuite.com/theory. And now, let's get back to the show. For the first time in history, companies have two different options for what to do when they have an open role that requires intelligence. That doesn't mean humans are done for, but what it does mean is that the bar for what a human needs to bring to the table to get selected when intelligence is what's needed has just gone up. And that means even if there are more jobs created, they will go to the most capable and cost-efficient option, human or otherwise. Even if there's an infinite appetite for intelligence, AI raises the floor of expected productivity and ability. And anyone who falls below that floor becomes part of what I'm calling the unproductive class or the charity class. That sounds horrible, but it's just the physics of the situation. AI learns slowly and narrowly, but it is hyper good at the tasks that it's been trained on. Full self-driving in cars is already proving to be far safer than human drivers. AI can beat us at chess and go in the most complicated games in the world. Now, that AI is never going to be a chef as well, but eventually, there will be a separate AI for that, too. And that's why this isn't just about companies downsizing, whether for PR or otherwise, it's about what happens to the structure of companies and society as a whole when not everyone will be needed in the workforce. What happens when one person can do the work of 20 or 200? The entire fabric of society is going to change. In an AI-enabled world, companies don't just get smaller, they atomize and become ephemeral, existing briefly and then completely disappearing, being replaced by something else, because the barrier to entry is so small. You get more companies, but with fewer people in them. They become lean, high-output, high-margin operations that previously would have required entire departments just a few years ago. So, now the question becomes, if one person can build a $401 million operation, bring on his brother, and achieve a run rate of nearly $2 billion in their second year of business, that's the pace they're on, and do it with better margins than their competitor that has nearly 2,500 employees, how long do you suppose those 2,400 employees actually last? It's not a rhetorical question. It's the question every upstart entrepreneur and company in America is asking themselves right now. And the inevitable answer that they will come to is that every company can now do more with less. That includes you. And that fact is going to change our economy for ever. Welcome to part three, the barbell, how the middle will disappear. America's already in the grips of a horrific K-shaped economy, where there is a stark divide between the haves at the top of the K, and the have-nots at the bottom. 10% of Americans own 93% of all assets. 47% of Americans already make so little money that they pay zero federal income tax. And despite what many will tell you, we're already a massive welfare state. As of 2022, a full one in three Americans were enrolled in at least one government assistance program. And now that the private sector job growth has stalled, you can expect those numbers to go up even without AI. And with AI? Expect those numbers to go up every time AI raises the bar on the acceptable level of productivity. Elon once said, "You're paid in direct proportion to the difficulty of the problems you solve." He's right about that. As AI becomes capable of solving increasingly complex problems, more and more humans will find themselves unable to obtain work. Unless they're willing to constantly push their abilities higher and higher, they will eventually find their careers stalled out or not existing at all. Now, some people love that kind of challenge, but others do not. And this divide between those two groups will further exaggerate the K-shaped economy as more people slide down into the unproductive class. AI will create almost perfect transparency with someone's level of productivity. There will be nowhere to hide. AI can parse through so much data so quickly and identify patterns that there will be absolutely no way for people to get away without delivering real productivity to the company. Even in a big company, everyone is going to be exposed. And this is going to cause workers to fight for regulatory protections. That will cause more entrepreneurs to avoid hiring humans in the first place. And from that, we will see companies begin to atomize dramatically. Corporations will develop a huge appetite for consultants and independent contractors, and more and more traditional employees will launch their own companies as a solopreneur to sell their skills back to companies that have a growing incentive to use AI or outside contractors whenever possible. And the people that master AI and are able to amplify their abilities, they're going to capture more of the upside of their own talents, something that up till now they've been unable to do, but that will radically increase this atomization. You will only get paid for your results in the new world. But odds are there will be less and less competition as you elevate your skills more and more. And that becomes the goal, to put yourself in that rarefied air. Eventually, your primary competition will be other elite performers using AI and then the AI itself. Everyone else is just going to fade away into the unproductive class. It is absolutely heartbreaking, but that is just the reality when intelligence has a distribution, drive, will, all of those things are distributed. Not everybody has the same level. Not everybody is going to be able to keep up with AI. That is the stark and biting reality. It creates huge opportunities for people that can and will, but ooh, buddy, society's going to look very different on the other side. Now, I know exactly what some of you are thinking right now because I read my comments. You're tired of hearing a bunch of rich guys like me who've made it stand up and tell you that the solution to declining jobs is to learn AI. You're watching real jobs disappear in real time, and the response from people like me is essentially, "Hey, it's great news. Just go get better." Now, I understand why that sounds so horrible, especially if you're of a certain age. But as Morpheus says, "All I'm offering is the truth." And the truth is, for the foreseeable future, AI alone is not going to beat humans. That will happen one day, but that day is not today. For now, what the data actually shows is that humans who use AI are absolutely demolishing both the humans who don't and the AI systems operating without human direction. And that gap between humans with AI and everyone else is the window of opportunity. It's open right now, but it will not be open forever. The barbell is going to form no matter what for the reasons that I've just explained. But if you've made it this far in this video, you meet what I call minimum requirements. Now, you need to start stacking relevant skills. If you do, you'll end up on the good side of the barbell. And if you don't, over the next 5 years or so, you will slide into the unproductive class as AI outpaces you. Remember, AI improves at roughly 300% per year. Even if that rate slows dramatically, in 5 years AI will be like a computer is today. You won't be able to get a job if you don't know how to use one. To be clear, I don't believe the middle class will disappear. The middle will fall out of the talent distribution. And that will create the barbell in productivity. The stable, salaried, not exceptional but competent enough middle, the people who always found a way to contribute just enough to justify their seat. That class is going to disappear because AI makes it optional for a company to carry them. And companies that figure that out will outperform those that don't, which means the companies that don't figure it out will eventually cease to exist. You're either productive enough to hire on merit, you have the skill and the gumption enough to run your own business, or you're going to fall back on charity. That'll be the divide. That's what the data is already showing us. So, welcome to part four. Now that you know, what do you do? The US economy is already producing zero net private sector jobs. AI is already absorbing the bottom of the cognitive labor market. Med Vi's success already proves that the atomization I'm predicting is certainly possible and probably likely. And on top of all of that, the Federal Reserve is caught in a trap it has never faced before. It can't cut rates right now to protect jobs without risking inflation. It can't raise rates to fight inflation without crushing an economy that is already producing zero net new jobs. Every lever that the Fed normally has was built for a different world than the one that we're in right now. Goldman Sachs, JP Morgan, and Bank of America are all currently signaling elevated recession risk. They might be right or they might be wrong, the future is very hard to see, but here's what's true regardless. The economy doesn't have to enter a recession for everything I've described not only to happen, but to accelerate. Slow growth will accelerate it. Margin pressure will accelerate it. A competitor who figures out the Med Vi model before you do will accelerate it. The forcing function isn't a single economic event. It's the competitive pressure that now exists in every industry, in every job, in every economic environment permanently. In every economic scenario that could unfold, strong growth, stagnation, or full-on recession, the pressure on every company will be identical. Do more with less. And to do that, you're going to use AI. In a strong economy, you compete on margins. In a stagflationary environment, you cut costs without cutting output. In a recession, your survival depends on eliminating inefficiency before the cycle turns. AI is the only tool that answers all of those demands simultaneously. The economic environment doesn't change the prescription. It just changes how urgently you must act on it, which brings me to what you actually need to do. First thing, if you've got the guts for entrepreneurship, now is your shot. I cannot stress this enough. Take control of your destiny. This moment is incredible for people that are willing to move, to push themselves, and to get better. It has quite literally never been easier to start a company than right now today. But if not, then at least stop thinking about your job as a job. Start thinking about it as a collection of tasks. Most people, when they ask, "Can AI replace me?" they're asking the wrong question at the wrong level of analysis. The right question is, "What are the six to 10 discrete things I actually do, and which of those can specifically be done better, faster, or cheaper by AI than I can do them right now?" Write that list down. Don't say it in your head. Write it down. On paper, in a document, break your role down to its actual components. Identify exactly where you're genuinely irreplaceable, where your judgment, your relationships, your creativity, where one of those things is the thing that makes you valuable. And also identify the tasks that are already automatable. If something is already automatable, it means someone above you is already thinking about automating that. I promise you, if they're not already, they will be soon. So, do not lie to yourself. Be very honest. And by the way, you can work with AI to think through all that stuff. The second thing, take the first automatable task on that list and spend a weekend actually trying to automate it. Not watching a YouTube video about it, not reading about it, actually attempting it. One tool, open the tool, vibe code it, build an agent, whatever's needed. Then run it, see what happens. It will probably fall on its face at first. Try again, improve it. The feedback loop from one real attempt to automate a task will teach you more than 20 hours, 40 hours of content consumption, and it will tell you exactly where the gaps are, which is where your human judgment still matters. The third thing, start thinking like an owner of a workflow, not an occupant of a role. Your job is to deliver outcomes. It's not to check a box like you're on a fetch quest. The workers who end up on the right side of the barbell are not going to be the ones who used AI to do their job a little bit faster. They're going to be the ones who used AI to become a department of one, to be the person who shows up and delivers what used to require a team. That's the person who gains leverage, whether that means charging more as an independent operator, or being genuinely irreplaceable inside of an organization because no single hire can replicate what you can deliver. Whatever it takes, get to that point. Prove that you can deliver a level of productivity that other people cannot, that AI by itself cannot, that truly you, armed with AI, have no peers other than people like you that have also mastered AI. None of this is complicated, but it does require you to actually start. So, here's the bottom line. This isn't a tech story. It's an economic structure story. Fed Chair Powell confirmed we have zero net private sector job creation. The job market's already stalled. Stanford has confirmed that entry-level cognitive work is already being absorbed by AI. Med Vi's success confirms that the atomization model isn't just theoretical, it's already happening, and it's incredibly profitable, but only for the people that can take advantage of it. And guys, anything that's profitable, especially on the scale of Med Vi, is going to be replicated over and over across every industry. And on top of all of that, the economy's struggling right now, at least for anyone on the bottom of the K. So, do not allow yourself to be on the bottom of the K and the wrong side of the barbell. That is far too brutal on the human psyche, and it's entirely avoidable. I'm not saying this moment is fun. I get it. We all wish that things were more static, that we had longer time horizons, but we don't. But we have this incredible opportunity. It is a window of opportunity that is open right now. Humans plus AI still beats both humans and AI alone. The people who move now will be ahead of the curve in terms of building up your AI mastery, and that will give you leverage and optionality that will compound for the rest of your career. But if you wait, you will wake up one day and find that that window of opportunity has become a wall. So, don't wait. Go master AI. All right, if you want to see me explore ideas like this in real time, be sure to hit that subscribe button right now. Join me Monday, Wednesday, and Friday at 7:00 a.m. Pacific time where we go live and talk about all of this stuff. You can join in the community and debate or just hang out and listen. Till then, my friends, be legendary. Take care. Peace. If you like this conversation, check out this episode to learn more. Oil prices have a much bigger impact on your retirement savings than you might think. And a lot of people are learning the hard way. The volatility in the Middle East means volatility in your portfolio. The last