How Big Tech Became The "Bad Guys"

It might be hard to remember today, but there was genuinely a time where the tech bros were widely considered the good guys. They were all making a lot of money, but they were still a welcome alternative to the tight financial industry that may have looked far more professional in their tailored suits, but still managed to blow up the global economy. By comparison, the tech industry was a laid-back counterculture, actually producing real tools and gadgets that were visibly making our lives better and more convenient with every iteration. For the people that worked for them, going into tech was a viable high-end career that still prioritized a work life balance and making the office somewhere that people actually wanted to be. And well, for the most part, the companies that had come to define this industry were also defining themselves as friendly, benign players that didn't take themselves too seriously and by definition weren't up to anything evil.

But well, that has clearly changed. Obviously, the public trust in these companies has shifted considerably, but the industry as a whole has not done much to push back against this shift. And if anything, they have leaned into it. Outside of the usual suspects like, well, whatever Palanteer is up to, even previously friendly faces like Google have senior executives openly talking about the destructive power of the tools they are working on. For their workers, they have gone from the best places to work in the world to a grindfest where 80-hour weeks have become the expectation. And for society at large, well, these are apparently smart people, or at least people that can afford a public relations team. So, they had to know this didn't look good, right? So, when exactly did big tech give up on being the good guys?

"If this technology goes wrong, it can go quite wrong. And we want to be vocal about that. We want to work with the government to prevent that from happening. But we we try to be very cleareyed about what the downside case is and the work that we have to do to mitigate that."

"Is the guy with the pentagram and the holy water and he's like, 'Yeah, he's sure you can control the demons.'"

"But when he's not in the office, he's usually at home on Facebook talking to all of our employees about different things that we're working on."

"And you'd have to search far and wide to find employees who get a better deal than Google's."

"Your life and mine will witness times where where there will be 20, 30, 50% unemployment in certain sectors. It's the rise of AI in an age where humanity is at its lowest morality."

"Shape the preferences and behavior of a powerful artificial mind such that it does not kill everyone."

Their peak tech bros enjoyed endless career progression opportunities, options packages that can make them multi-millionaires before they were 30, a good work life balance, and the promise that if they knew how to code, they would always be guaranteed a good job. Today, they are being forced back into the overpriced cities that don't want them there. They are being laid off in mass and the ones that remain are being told to grind. And they have apparently made programs so good that their CEOs are gleefully talking about a future where entry-level programmers are completely replaced by AI.

Between the golden age and the dark ages of tech bros, there have been three major changes that both caused their massive rise in their current fall. The first problem was that they simply ran themselves out of business. On Friday, the 10th of March in the year 2000, the NASDAQ hit its then all-time peak. Investors were all in on companies that were ready to take advantage of the internet. And the speculative mania was so fierce that just the mention of "dot" in their company filings was enough to send their stock price soaring. Eventually, this all collapsed, and only a select few companies from that era have survived to this day. The reality was that the internet was an amazing technology that would go on to change our lives and facilitate some of the most valuable companies in history. But just putting a normal business on a website wasn't actually revolutionary.

What came after was a genuine golden era of tech services. The companies that survived the dotcom bubble were actually useful and the cleansing of the market meant that more talent and investment could go towards supporting companies that actually did provide a good service. Early YouTube, MySpace, Facebook, Amazon, and online gaming were all novel and genuinely great services that if you are as old as me, you probably remember finding them absolutely amazing. There was still money flowing into them and the tech sector wrote out the global financial crisis better than most. The first iPhone was released in 2007, and even at the height of layoffs and corporate bankruptcies in 2008, people were still lined up around the block to buy the iPhone 3G. You were told that if you learned how to code, you would be sure of a good job. And at the same time, companies like Apple and Google were being named as the most desirable places to work thanks to their relaxed corporate attitude, excellent work life balance, amazing employee perks, and surprisingly competitive salaries. Most importantly, these companies were still not the establishment. They were scrappy startups making products that people actually enjoyed.

After the .com crash, it took the NASDAQ over 15 years to recover from its previous peak. This meant that thanks to employee stock options, these mid-2000s tech bros were making good money. But they would likely still be out-earned by their finance bro peers, assuming they weren't laid off in 2008. That slowly changed though, and as these companies grew, more money flowed into the industry. And with more money came more workers, bigger bonuses, stock options, and it went from an alternative industry filled with nerds making products that people enjoyed to the industry making products that would secure funding from a growing pool of investors. They went from the business of adding value through technology to extracting value through monopolies.

For a while, you could have a great degree of confidence in becoming filthy rich by putting in a few years at a major Silicon Valley tech company. But all this relied on a stream of money that wasn't coming from nowhere. Venture capital, the firms that actually invest in early-stage startups to develop their new technology, never again actually reached the level of funding it did during the dotcom bubble. That was until something changed in 2021. So, it's time to learn how money works to find out how tech bros ruined tech for themselves.

This video is sponsored by Monarch Money, which has basically become my go-to place for keeping track of my finances. Monarch is private and ad-free. It syncs with over 13,000 banks, credit cards, and investing accounts, and pulls everything into one clean dashboard. No ads, no data selling. You're the customer, not the product. In Monarch, I can see my spending, net worth, debt payoff, recurring bills, and even my credit score all in one place. The budgeting setup is flexible with simple rules and categories that keep things organized without feeling strict. And Monarch quietly combs through your numbers and surfaces patterns like spending changes or unusual transactions explained in plain English. There is also a built-in assistant you can ask anything about your finances, like how much you spent on travel last summer, and it answers from your actual data, not generic advice. I use it during my monthly budget check-ins, and having everything synced into one view makes things feel way calmer and less overwhelming. I can get the full picture with a quick glance instead of bouncing between apps. If you share money with a partner, Monarch has a shared view so you are both looking at the same goals, bills, and overall financial picture. If you want your finances to feel more clear and under control, scan the QR code on screen and use code HMW50 for 50% off your first year.

As tech companies grew larger and more valuable, the rush to hire more talent heated up. By the mid-2010s, if you had a talented developer, you could easily out-earn any other career out of college by working for a company like Facebook, Apple, Amazon, Netflix, or Google, the so-called FANG companies. These businesses all had established platforms, but they still hired tens of thousands of developers and paid them very well for a few reasons. The first was that they were constantly trying to add new features to their core offerings. And even maintaining a platform like YouTube takes a lot of work to keep up to date with customer demands. The homepage of the site today looks very different from even 5 years ago.

There was also an exuberance where newer companies like Uber, Airbnb, Twitch, Snapchat, Tinder, and even WeWork were scaling their operations so fast that they would hire developers before they even had a defined project for them to work on because hiring when they actually needed staff would slow them down. This was part of a relatively new tech industry strategy called blitz scaling. A not-so-subtle reference to the blitzkrieg of World War II, which was all about capturing as much territory as possible as quickly as possible using new technology before the enemies could respond. Supply lines couldn't keep up. But the hope was that once a territory was captured, the German army could sort all of that out later. Now, clearly tech companies are not rolling tanks through Belgium, but they are using the scalability of technology to capture market share in industries like food delivery, taxis, holiday rentals, and online dating as quickly as possible and then sorting out things like what to do with all their staff later on. In an interview published by the Harvard Business Review, Reed Hoffman, a Silicon Valley venture capitalist and one of the founders of PayPal and LinkedIn, said that companies that are blitz scaling may need to get as many warm bodies through the door as possible as quickly as you can.

Another reason why tech bros are being hired as quickly as possible was that hiring lots of staff was a great way to make sure that they couldn't go to potential competitors, especially startups that could, and I I really hate to say this, but disrupt the industries of the big incumbent players. Acquiring competitor companies as they start to take market share could get companies in trouble with the FTC for anti-competitive practices, but there were no rules against denying them staff that they would need to get their business going in the first place. A report by the Wall Street Journal interviewed tech workers who admitted to being hired to do nothing at all and that the businesses were hoarding developers like Pokémon cards.

That made it all worse when high interest rates, reduced market certainty, and lower customer demand for a lot of these tech products resulted in mass layoffs across the industry. Businesses that were blitz scaling either shut down or shifted gears to only hire workers they really actually needed. New feature development slowed down and big established companies didn't need to worry about hoarding workers away from their competitors anymore because nobody was hiring. Artificial intelligence is also already doing a lot of the grunt work that is typically given to new employees. Tech companies by their nature are more open to adopting these new technologies and even replacing just a handful of entry-level developers can save companies millions of dollars every year. So, it really is an ideal environment for automation.

This was the second major blow to the tech bros. They traded in the promise of a very stable career for something that has become an incredibly risky game of survival. You can make a lot of money in technology, but your success is going to depend just as much on picking the right company or startup at the right time to vest in as it is on your own personal skills development. An entry-level developer that started working with Intel 5 years ago had a good chance of being laid off last month with stock options that are worth absolutely nothing. A similarly skilled developer that took a job with Nvidia at the same time, on the other hand, probably never needs to work again in their life. The ones that got lucky go out. And the ones that got let go are sick of playing the stock market with their careers. If they wanted to do that, they would have become finance bros.

And then there is the biggest problem of all. Workers want to be close to job opportunities and companies want to be close to workers. So the whole industry has agglomerated in just a handful of locations. These cities have become incredibly expensive, and infrastructure has not been able to scale as fast as these companies, which means it's not unusual for tech workers making six figures to share a two-bedroom apartment with three other highly paid local workers. Ask me how I know. If you are a tech worker struggling to afford a place to live on a six-figure salary, then there is basically no hope for other workers in these cities, which are still essential to keep society functioning. As a result, big tech cities have not only become expensive places to live, they've also become not particularly nice places to live. Remote work could have alleviated these issues by easing demand in these small markets. But the big tech companies now have a lot more negotiating power over workers who don't want to be another layoff statistic, and they are using that power to mandate a return to office. Even Zoom, the company developing remote working solutions, has pulled its teams back into the office.

These problems also don't come with many upsides for locals. The big sell is that these companies offer great jobs, but most of these jobs go to people who are not local and only move to the city after they secure a role and then compete with locals for housing and other services. I personally moved to San Francisco from out of state after I got a job at an investment bank. So even though I was technically a finance bro, I was still very much a part of the problem. Other areas have gone from welcoming tech companies opening up operations in their cities to actively pushing back against it. And this all represents a trend which is way larger than just gentrifying entire cities. People just straight up don't like the tech industry anymore. Whether they are pricing them out of their homes, collecting their data, hitting them with declining services, making them piss in bottles, or threatening their jobs with AI, most people don't see this as progress anymore. And the tech bros have become just as unpopular as us finance bros.

And the reason for that isn't just the layoffs or the poor working conditions. It's that a lot of these companies built their empires by simply deciding that the rules don't apply to them. Big technology companies naturally want to stretch the limits of what they are allowed to do within the confines of the law. But effective regulators are supposed to keep them within these guidelines. The only problem is that there is much, much more money to be made on the other side of the technical word of the law. So over time they have developed a foolproof system for getting around these legal suggestions to access markets that nobody else thought was possible. Uber got around taxi regulations, food delivery apps, minimum wage laws, buy now pay later services, consumer lending standards. Airbnb got around hotel licensing and zoning laws, crowdfunding investment standards. And then there are even more obvious examples like bet-on-everything platforms that have skirted gambling regulations, not to mention age limits based on extremely flimsy financial technicalities. This isn't even touching the absolute dumpster fire that has been allowed in the gray areas of cryptocurrency. But the thing is, every single one of these companies that bravely chose to venture where the law said they couldn't are all worth billions of dollars today. In the battle between technology and the law, technology is winning every single time.

In the short term, there are actually some genuine benefits to workers and consumers by breaking down rigid structures. But if they were willing to bend the rules in our favor to get us on board in the first place, it sets a pretty bad precedent once they have captured a market, which is exactly what we are seeing today. Now, the solution is effective, well-funded, and well-staffed regulators to make sure that these businesses are abiding by the word of the law from day one. But come on, let's be honest with ourselves. We all know that's not happening anytime soon. So instead of that, the best way to see this problem in action is to go through the process step by step. So that if you yourself want to build a billion-dollar tech platform by skirting the word of the law, at least you will have the tried and tested game plan for doing it right.

So this business strategy isn't actually anything that new by itself. It's called regulatory arbitrage, and it's been a tool for opportunistic businesses for as long as rules have existed. The idea is that if you can find a loophole or gray area in the word of the law, you can bring a more competitive product to market than your competitors working within the spirit of the law. The term itself gained popularity in the 1980s when the Basel Accords attempted to introduce a series of best practices for the global financial system. Banks in different countries operating under different central banks had different capital requirements, which in plain English meant that they had to keep a different amount of their total deposits on hand instead of lending them out. In some countries, banks only had to keep 5% of their total deposits on hand, whereas in other countries, they had to keep as much as 25%. The banks that had to keep more cash sitting around were safer. But because they couldn't lend the money out or invest it, they couldn't offer globally competitive interest rates, and they would naturally make lower profit. They figured out that by just having operations in the loosest legal jurisdictions, they could arbitrage the difference to gain an advantage over other banks operating to the standards of that market. Ironically, Basel 1 and subsequently Basel 2 and 3 have all been international agreements on banking regulation to avoid this kind of behavior. But a lot of the financial system responded to these international baselines by doing more regulatory arbitrage. One trick was to open mortgage originator or non-bank lender operations that were funded by the bank, advertised by the bank, and managed by the bank, but weren't technically part of the bank. This let them write more mortgages for a given amount of capital than they would have otherwise been allowed to. And well, we all know that eventually turned out super well.

But today, this little game of "find the loopholes before they get plugged" has become much easier and even more profitable. So, if you want to get in on this action, step one is to find an industry with a lot of regulations that make it kind of terrible. For our little company, let's go after schools. People generally don't like the school system, but competing normally in this market is kind of difficult because there are a lot of regulations around how these institutions are supposed to be run. These add costs, complexity, and rigidity that you won't need to worry about if you are able to follow through on the rest of this plan to build a cloud-based, decentralized, gig-based school system where freelance teachers can earn commissions for teaching classes.

The next step is to find a non-commercial version of your industry that is kind of similar enough to give you the "don't worry about it, bro" legal defense. For Uber, it was "don't worry about it, bro. These aren't fake taxis without proper registration. They are people carpooling together." For Airbnb, it was "don't worry about it, bro. These aren't hotels operating outside of licensing and local zoning laws. These are just friends who met online a day ago, becoming really short-term roommates." For "buy now pay later," well, you guessed it. "Don't worry, bro. These aren't unregulated predatory loans made without lending licenses or standards. This is just store layaway handled online for you in your new platform." Let's call it something insufferable like Schoolie. You should go with the line, "Don't worry about it, bro. This isn't an unregistered school with unqualified teachers. It's a network that lets parents share the burden of homeschooling with one another with some fees to cover the cost of running a class."

Now, this strategy won't work forever, but it doesn't need to. Modern network systems have become the secret ingredient that you will need to turn this flimsy argument into a multi-billion dollar rival to the outdated and inefficient incumbents. Okay, so step three is to make sure your new regulatory loophole leverages technology wherever it can. And there are three reasons why this alone is so important. The first is that technology like user networks can scale much faster than handling the entire business yourself. Once you start exploiting a legal loophole, the clock is ticking before regulators catch up to you or lawmakers simply plug the gap. Yeah, sure, none of these groups are famous for moving particularly fast, especially if you throw a few campaign contributions their way, but in this game, they are sort of like that killer snail. Get complacent for long enough and they will end you. If you had to individually build all of these decentralized schools, hire all of these teaching consultants, and onboard all of the students, you wouldn't make it far enough through the rest of the game plan before the legal problems caught up with you. Technology-enabled networks let you just build a system for everybody else to do the hard stuff at scale, letting you move much faster.

The second benefit to just building the technology is that it is much cheaper in the long run. It's better to let someone else worry about maintaining their pre-10 taxi or pretend hotel or pretend school. You just want to be a convenient middleman. The third benefit is that scalable technology is going to have a much easier time raising money from venture capitalists who are willing to make big gambles like this. Developing your own school takes a lot of cash that may never get returned. And even if you can run a profitable business in such a strange market, those returns will be slower. The same amount of money spent building a platform that enables other people to build schools can theoretically scale as quickly as the servers can register new users. So the potential for thousandx returns is greater, which is important to these venture capitalists who want to cover the losses from 999 similar businesses that didn't make it.

So once you have found your loophole and raised money to build a piece of technology to exploit it quickly and efficiently, it's time for step four, which is to launch it. But the trick here is to be really careful about where you do that. Certain cities, states, and countries are just way more chill about this sort of stuff than others. You want to shop around for a government that is the least likely to cause you problems for this first launch. Historically, places like Silicon Valley or Austin have been popular choices because they are generally more supportive of technology. But these days, you shouldn't be afraid of looking further abroad. For Schoolie, your best bet is probably going to be a small city with a terrible school board, underperforming students, and a state government amicable to home and charter teaching.

Once you have selected a location or a handful of locations for your initial launch, step five is to do absolutely anything it takes to make these early rollouts massively successful to all stakeholders involved, except for your shareholders. For now, feel free to burn through investor funding to provide wildly subsidized services. Pay your gig teachers double what they could earn in a regular school. And also make sure the wider community appreciates your business as well. Fund sporting events, support local charities, talk about all of the bureaucratic problems you are solving, the taxpayer money you are saving, and how great it is that Nowhere'sville, West Missabama, is beating the big cities by embracing the future. Now, ticking all these boxes will likely cost your business a lot of money. But really, this is just a marketing expense to buy some political ammunition down the road because step six is to repeat the formula everywhere. Basically, you want to search low to high on cities by regulatory oversight and work your way down the list as quickly as possible. Some of these places will inevitably challenge you, but hopefully by this point, you have raised enough investor cash that you can drag the underfunded and understaffed local regulators through the court for years before it becomes a real problem. All the while you're gaining new users and contractors every single day.

Step seven is to become too big to nail. If you have scaled your business properly, by the time the courts have made any progress, you will be serving millions of users and employing, oh, sorry, I mean independently offering business opportunities to hundreds of thousands of contractors. Uber is now practically one of the largest employers in the world. Even if it doesn't technically employ any of its drivers at this point, even if it was absolutely nailed in court for clearly violating rules around operating unregistered taxis, it's too late. Politicians will simply not allow that many people to suddenly show up as officially unemployed on their watch. If Schoolie can offer the same opportunities to its homeschooling specialists, as well as sprinkle in a bit of "won't someone please think of the children," then once it gets to a certain size, it's very unlikely that any politician will have the grit to challenge you. Now, once you have all of these pieces in place, feel free to "monetize" your platform, squeeze those margins, and get to work paying back all of the investors that funded this process in the first place. In fact, a lot of those early venture capitalists will help guide you along this checklist because they do it all the time and they know it's the best chance of them getting their money back.

Now, it sounds pretty bad, right? But just to mix things up a bit, it gets better. Sometimes this whole process is actually exactly what a market needs. The opposite side of regulatory arbitrage is regulatory capture. And if anything, that can be even more damaging. This is where incumbent businesses intentionally layer on mountains of red tape and regulations that are extremely difficult for new market entrants to deal with and in extreme cases actually effectively outlaw competition altogether. Nobody has used that playbook more effectively or squeezed more out of it than Amazon.

The tech conglomerate has so many fingers in so many pies it has found itself exposed to basically every risk and opportunity in the market today. Softening demand for maxed-out customers, pushback from exploited third-party retailers, returns fraud, investors who want to actually start seeing some profits on all the AI hype, a growing dependence on advertising revenue, a degrading customer experience, competition from Chinese retailers who are selling a lot of the same crap anyway, and the fact that they are literally running out of human beings who haven't already worked for them before and quit. It's easy and honestly sometimes a little bit fun to point out the flaws and challenges in a business that has screwed over so many people. But the Amazon case study is more important than that. It perfectly encapsulates the strengths and weaknesses of the modern business strategy as it tries to be so many different things to so many different people. You might think that all of these problems are just another recession indicator or yet another case of juicing short-term gains at the expense of long-term reputation. But the inconvenient truth is that a lot of these challenges were inevitable from the very beginning.

Online retail is the most competitive business in the world. Amazon dominated early e-commerce by offering an incredible service below cost to build a one-stop marketplace. But now that it has, it cannot maintain it without making a long list of very unpopular sacrifices. Despite all of the problems in the world, the American spirit to consume is still going strong. This year, Amazon extended its Prime Day sales to 4 days and once again announced record results. The expectation was that the company, in combination with the other retailers that were forced to put on sales in response, would generate two Black Fridays worth of sales over this time, just without the trampling. And I mean, really, what's the point of that? But behind the scenes, there was cause for concern. Their official announcement focused more on how AI was helping people shop and how much people saved. But nowhere did it list how much they had actually spent. Third-party guesstimates have said that despite running for twice as many days as last year and offering longer overall savings on big-ticket items, sales growth was only up by 4.9%, with other reports suggesting results significantly worse.

Reading between the lines of the company's press release, they were focused almost exclusively on the service benefits that come with a Prime membership like Amazon Music, fuel discounts, a GrubHub subscription, and Prime Video, rather than the retail side of their business because that business kind of sucks. Amazon has invested tens of billions of dollars to build out the most efficient, hyper-optimized distribution network in the world. It has massive economies of scale, petabytes of customer data, a captive market of subscription customers, and even then, the actual retail operations of its business are barely profitable. Express delivering individual packages to customers' doors is just really expensive when compared to a competitor like Walmart, where customers have to take themselves to the products, not the other way around. The point is the company just doesn't have that much wiggle room when it wants to compete on price with a fundamentally cheaper system. So it has had to slowly make sacrifices which are admittedly looking excellent on financial statements but have caused four serious problems in a business that markets itself on customer obsession.

The first problem is a growing reliance on advertising bloat. Amazon is now one of the largest advertising businesses in the world. According to company financials, they have made almost $60 billion last year from advertising alone, a revenue segment that outpaced every other part of the business. Amazon now brings in more money from advertising than it does through its Prime subscription service by a significant margin. The most publicized place where these ads started popping up was on Prime Video. The company was one of the first to roll out ads on top of an already paid streaming service. For the three of you who have actually watched Rings of Power, I am sure this was very annoying. Now, it would be hypocritical of me to say that advertising is just always bad. I wouldn't be able to pay my team or myself if it wasn't for advertising. And I very much believe that as long as it is reputable, I would rather take an advertiser's money than my audience's. However, when people were already paying for a service, there was understandably a bit more pushback. But stream advertising wasn't where the real money or the real problems were. Anyway, if you had tried to search for a specific product on Amazon recently, I don't need to tell you that sponsored listings have become far more invasive. This advertising space is some of the most valuable anywhere in the world because people who are literally searching for a specific product in an online marketplace are about the most high-intent customers imaginable. Brands are now spending several billion dollars per month to get their products to show up first. If the company can't sell ad spots in a particular category, they will just give them away to their own white label brands. There also isn't a lot of pressure from the company to make sure these advertisers are promoting products that are, well, you know, good. An investigation by the FTC found that Bezos himself, while he was still in charge, pushed executives to accept junk ads because they were so profitable that in a report he was quoted saying, "The revenue generated by advertisements eclipses the revenue lost by degrading consumer shopping experience." Now, to be fair, he was right. Over the last year, Amazon's combined retail operations have grown their margins to be wider than even Walmart, who you will remember has lower operating overhead. They might talk a lot about high-tech automation in their warehouses or AI shopping assistance, but the real driver of these profit margins has been advertising within the online store itself. This has fundamentally changed the nature of the business. The company is still customer-obsessed, but you are not their customer. The customers are advertisers. You and the millions of other people that shop online through them every month are the product they sell to those customers. And that's just the first problem.

So, you are probably already aware that for most of Amazon's history, the loss-making storefront was subsidized by early investors and then the company's massively profitable Amazon Web Services. Amazon was one of the first to realize that it could effectively rent out the spare space on their server infrastructure and let other businesses get access to globalized hosting without having to build out their own data centers. In the most recent letter to shareholders, they made special mention of recent contracts they have signed with organizations like PayPal, Norwegian Cruise Lines, Reddit, Japan Airlines, Northrup Grumman, and the US Army. These contracts will pay out billions of dollars with strong margins because once the upfront investment in the server hosting is made, the actual ongoing expenses are comparatively minimal. This part of the business makes more profit every year than the company has ever made from retail. Which begs the question, why don't they just give up on retail which is causing them so many problems and focus exclusively on AWS? Well, that has a lot to do with the second major problem Amazon is facing. The uncertain value of all that money they are throwing into AI.

In the last annual letter to shareholders, the company CEO spoke about next generation's "wise" in their business. And the top three "wise" that he was addressing from shareholders was: Why is AI so important? Why invest so much into AI this quickly? Why do chips and AI need to be this expensive? And why have personal assistants not taken off yet? He gave the standard corporate-approved answers and it's a public release, so I am not going to waste your time reading them here. But the questions themselves highlighted some pretty clear pressure from shareholders to actually see some results on all the promises of AI. According to Bloomberg, top tech companies spent $200 billion last year on capital expenditures related to AI. The biggest investor out of all of them was Amazon, which alone sunk more than $70 billion into AI infrastructure. To use the technical corporate terminology, that is a ton of money. But it's actually worse than it sounds. This is just capital expenditure, which typically means money spent on tangible assets like office space, warehouses, trucks, and in this case, racks on racks on racks of GPUs. That means it does not include the salaries for all the developers, the R&D budget to develop their own chips, the terawatt hours of energy they are pumping into this, or the missed opportunity of what could have been done with this money instead. Capital expenditures also typically build up capital in the business. If a company spends a billion dollars on a new office, they haven't lost a billion dollars. They've just transferred it from cash into real estate on the balance sheet. But the Nvidia chips that make up a majority of Amazon spending have a very short shelf life. In 10 years' time, a commercial real estate investment will still be worth money. If the business is lucky, it might have even appreciated in value. But to those of you who are still running a rig with a 1060 in it, you will know that in 10 years, these current state-of-the-art GPUs aren't going to be worth diddly. Setting money on fire at this scale has literally never been done in the history of business. And there are a lot of shareholders that are suggesting that maybe it would just be better to do stock buybacks with that money instead. However, if this all pays off, Amazon's retail division would stand to benefit more than any other business in the world. Automated shopping, warehousing, and last-mile delivery would transform their current high-revenue, low-margin business into an even higher revenue, high-margin business. But with every year that goes by, there are tens of billions of dollars that need to be made up with these cost savings. And that much money pays for a lot of wage cages and, well, let's call them tactical jugs.

But that's actually the third problem, which also explains why Amazon is going all-in on AI and automation. They are running out of people to work for them. Amazon is the second largest private employer in America after Walmart, and most of that headcount is warehouse or delivery staff. The conditions in these roles are very bad. Warehouse and delivery work is physically demanding. The pay is low and the deadlines are strict and turnover is massive. According to leaked documents published by Engadget, the company had a turnover rate of 150%. Which in plain English means it was effectively replacing its entire workforce every 9 months. With more than a million employees in total, this means that it would only take about a decade before literally every abled-bodied working-age person in America who wasn't already employed at a better job would have worked for Amazon. Even aside from the harsh working conditions, the company does not encourage employee tenure, statistically preferring to hire new college graduates over offering internal promotions to warehouse staff. Only around 4% of warehouse managers, let alone floor staff, were promoted into a corporate role. According to the same report, the company could pay its employees more or relax the punishing deadlines, but that would either cost them the margins they had fought so hard for or slow down the fast deliveries that they have built their brand on. The hope that they have been relying on for almost two decades at this point is mass automation of the retail operations that will eliminate the need for masses of floor staff in the first place. In the meantime, even though retail doesn't actually make them that much money, it's a hand that they get to hold on to for free, which could pay off massively if their dream of comprehensive end-to-end automation comes true.

There are also some benefits of having lots of revenue, even if it doesn't necessarily translate into lots of profit. Amazon makes billions of dollars in sales every day, but it doesn't need to pay for the items it's selling for up to 3 months after it has already sold them. This means that the company gets to hold on to a whole lot of cash in the meantime, which is something called negative working capital. In the case of Amazon, though, this is billions of dollars, which helps with a lot of day-to-day operations. But it is an additional strain on the fourth big problem they are contending with.

Third-party sellers hate using Amazon. A lot of the products the company sells, they do not own. They are just distributed on behalf of third parties through the Fulfilled by Amazon or FBA service. A lot of businesses selling products online feel forced to list their items on Amazon because that's often the first stop for people that want a convenient online shopping experience. But they would really rather not for three reasons. The first reason is that if they are selling generic items, their products are all mixed together with other products with the same skill, even if they are from different suppliers. This has made counterfeit products extremely common because once they are in the distribution system, there is no way to tell which third party was responsible for the defective items, but whoever happened to take the order will be blamed for it. The same problem happens with returns. Amazon has a growing problem with people taking advantage of its generous returns policy. Some people take this to an extreme by systematically cycling through Amazon accounts to try and get expensive items for free. Now, I'm sorry, but I am not going to teach you how to do this, but it generally involves buying a product and then getting that money back without actually returning it or not returning it in a good enough condition and to go and buy it at a discount later. I know I sound like a broken record, but a lot of people in America and around the world right now are maxed out on debt and bottomed out on savings. If people can make a few hundred bucks pulling a fast one with some online orders, some are going to do it. When the perceived victim is Amazon and they don't even need to hand back the products in person, people don't feel as guilty about returns fraud as they would from an independent business. Even when most of the time it actually is the third-party retailer behind the scenes. Since items are hand-delivered to your door, it's often not even worth it for the company to get someone to come and pick an item up again because then they are paying for shipping twice on an item they probably won't be able to sell again. Every year, the company destroys billions of dollars worth of returned merchandise, which the third-party retailers end up paying for. According to data from Red Stack Fulfillment, this is lower than most e-commerce sales, which average a return rate of 16.9% compared to Amazon's 5 to 15%, which it said was thanks to Amazon's investments in customer ratings. However, when businesses handle their own returns directly, they have more control over what happens to the product, which can avoid new customers getting defective repackaged returns. This return rate is also almost double the rate of traditional brick-and-mortar retailers because getting away with questionable returns is a lot harder in person. For businesses that sell through Amazon, this is a growing strain on their business. And for normal people that just want to return defective items, they now have to jump through more hoops than ever to get it done.

Between customers, shareholders, employees, third-party retailers, and economic realities, Amazon cannot satisfy everybody at the same time. For now, the strategy is to make everybody just a little bit angry while they wait for AI to hopefully pay off. But even if that does, it presents a new problem. If AI takes all of our jobs, who is Amazon going to sell to? Whether we like it or not, our immediate economic future is riding on the success or failure of this technology. Just the top companies are on track to spend $400 billion on data center buildouts this year alone. And the entire stock market has clearly been buoyed by investor hype around what this technology could do, at some undetermined point in the future. Now, that has done two things in the short term. Trillions of dollars worth of data centers may or may not turn out to be a spectacular waste of money, but at least for now, they are providing tens of thousands of high-paying jobs to engineers, tradesmen, and technicians all across the country. The same goes for the stock market. These impressive gains may or may not be backed by sustainable fundamentals. But at least for now, people who have seen their portfolios double in value over the last 3 years are feeling pretty good about treating themselves to a bit of conspicuous consumption, helping to keep the rest of the economy afloat. Now, I know the problem of our entire economy being kept alive by unsustainable capital expenditure and questionable financial gains is not exactly shocking news to anybody anymore. In fact, if you're watching this video, there's a good chance your entire subscription feed hasn't shut up about the AI bubble all year. And in our defense, if this bubble pops, it will probably take the rest of the already shaky economy down with it. This also makes the fact that people are treating it like a foregone conclusion all the more dangerous. But perhaps we have all been missing the more important question, which is what does a good outcome actually look like at this point? Is there any way that we could still quietly back out of this unscathed?

Okay, so to really hedge our bets, we need to look at the three ways this technology could pan out and what each of them would do to the economy that regular people live in. The optimistic outcome where artificial intelligence becomes exponentially more capable and comprehensively substitutes human workers in almost every profession. The pessimistic outcome where technology plateaus and it becomes clear that trillions of dollars worth of data centers were a huge waste of resources. And perhaps the scariest outcome of all, which is a scenario where this technology finds some useful applications, but overall just continues to be a bit meh.

So, the optimistic outcome is what most of the big AI companies and personalities are pitching to the world. That is, with the exception of Peter Thiel. I am not really sure what he wants.

"You would prefer the human race to endure, right?"

"Uh, you're hesitating. Oh."

Anyway, the idea of this scenario is that AI models and the tools they enable get so good that it renders most human work obsolete. And whatever human work is left becomes thousands of times more productive thanks to tools that do the non-creative work for us. Skilled trades are done by highly capable humanoid robots. Logistics is fully automated and even health and elderly services leverage mechanical muscles and mechanical minds to care for an aging population with a smaller group of actual doctors watching over everything. Eventually, the only jobs left will be those that need human creativity, assuming we can't automate that, too. If you have always wanted to design your own video game, you won't need a full development team. You could just "vibe code" your way through production with AI tools, creating something that used to take thousands of real man-hours. Yes, I know a lot of you actual developers or skilled artisans are rolling your eyes right now. But remember, this is the dream scenario the AI companies are selling. And to

Play devil's advocate, it's kind of already happening. Even if you completely ignore the current focus on generative AI and robotics, there are hundreds of technologies that have already done the same thing. To use the example of making your own game again, there are modern development platforms and tools that have made it far easier for a single person to put together a pretty good game, even working out of their bedroom.

The same thing has happened across almost every industry in our economy. It's the reason why our labor productivity gets better and better almost every year, meaning for every hour that we work, we are producing more value. The optimistic view of new AI technology is that it will accelerate this trend so much that just a few hours of human labor, mostly to oversee the work of the machines, will produce enough value to give us a lifestyle we couldn't even dream of today. So yeah, even if we lose our 9 to 5 jobs, it won't really matter because there will be so much stuff getting made that only a minuscule amount of hours will need to be traded in return for everything you could ever need. In fact, for a lot of people, it could be so minimal that it's not really worth having any kind of formal employment at all.

Everybody from Sam Altman to Jerome Powell has said the same thing about AI improving productivity, although admittedly with differing degrees of optimism. And the thing is, the idea itself isn't inherently wrong. But there is a small problem. Worker productivity has already increased by a lot for a number of reasons, including technology. But for at least 50 years, almost none of that increase has resulted in higher compensation for regular workers. So then why would this time be any different?

Well, the big AI players actually do have some answers to this obvious question, but these talking points need to be considered alongside some logical ironies that would almost be funny if they weren't so important. So, the widening productivity compensation gap is one of the most hotly debated issues in all of economics because everybody has something different they would like to blame it on. Some people will say that migration has flooded the labor market with people who are willing to accept lower wages for the same job. Some people point to women entering the workforce, which also increased the total supply of workers. Some people point to outsourcing high labor industries like manufacturing or technologies like computer systems that made the same amount of clerical work possible with a far smaller team. Then of course other people will say it's because labor unions were crippled in the early 1980s which made it harder for workers to negotiate collectively and turned the jobs market into every man for themselves.

Now the reality is that it was probably a little bit of everything because they all effectively did the same thing. Businesses got access to a greater supply of labor than they demanded and without any negotiating power that pushed labor prices down. AI is really just another technology that will mean businesses demand fewer employees to perform the same operations and something that reduces negotiating power as well. A report published by the executive outplacement firm Challenger Gray in Christmas tracks layoff announcements and categorizes them by type. This year outside of federal worker layoffs from Doge, the number one reason cited for layoffs was a cost cutting shift towards AI. In total, since the firm started tracking this category of layoffs in 2023, more than 150,000 job losses have been directly attributed to this technology.

Now, I know what you're about to say. Yes, a lot of companies are just claiming that they are doing layoffs because of AI when really they just want an investor-friendly way of saying that they are cutting costs. But that's actually exactly the point. Even without the ultra-optimistic tools that could be made possible by this technology, it's already becoming a great negotiating tool against workers pushing the gap further apart rather than closing it.

So this is where we get onto the narrative of living in a world of such material abundance that the idea of money becomes completely irrelevant. It's a nice idea, but there are a few problems. As I was putting this video together, a group of researchers from OpenAI very publicly resigned because they said that the company's economic studies were slowly being pushed into outright AI advocacy. If they found something that would present a future of AI in a negative light, they would just bury it and broaden the scope of their research until the results looked good. One of them was quoted saying, "The economic research team was veering away from doing real research and instead acting like its employer's propaganda arm."

Now, this is just the word of one man, but he probably gave up a pretty big payday to get the message out, and it does align with other reports coming out of the organization. The technically separate organization, Open Research, conducted one of the biggest universal basic income experiments ever on low-income households. The results were mixed at best, but the headline press releases were overly positive about how great an AI-powered UBI future will be for everybody. Now, maybe it shouldn't be shocking that research coming out of a company with a very clear motivation to push mass AI adoption would be a little bit biased, but the ironies do go deeper than that.

Remember when OpenAI was a nonprofit and the founders didn't care about making money off it? Well, according to reports from Reuters, it is now gearing up to go public in a move that could net the CEO as much as $10 billion. Now, if only he still had any economists left on his team, they could tell him that all of that money will be meaningless in the coming AI utopia.

All right, joking aside though, it's pretty clear that even if this technology meets the expectations of even the most ambitious technoptimists, there are very few systems in place to ensure that the increased productivity from AI actually benefits regular people. It's a lot of technoeconomic jargon to try and gloss over what you probably already know, which is that if the people that own these tools can make a lot of money off not needing to pay workers anymore, they are going to try their very best to do it.

But that was the optimistic outcome. The most immediate fear that a lot of people have right now is that the hype around this technology will die off and bring down the entire economy with it. Again, as I was putting this video together, the jobs report for November was published after October was skipped due to the government shutdown. Overall, unemployment rose to 4.6% with an additional 4.1% of people in part-time or gig work who would prefer to be working full-time. This is the highest it has been since 2021, and it's clearly trending upwards. This is also before revisions which have consistently been downwards for every report so far this year.

It's not great, but the numbers have been helped a lot by three standout categories. The first two were healthcare and social services primarily to look after an aging and ailing population. The third category was construction, specifically specialized non-residential construction or in plain English all of those data centers. This was something that the BLS explicitly highlighted in its March report and has only become more relevant as the rate of data center buildouts accelerates and other government infrastructure projects are scaled back.

Now eventually this will be a problem because theoretically we are going to slow down on building these data centers at some point before we turn our entire solar system into a matrioska brain. So there's still like a lot of things like I would love to go build the Dyson sphere around the solar system and like you know make the world's gigantic data center with the entire energy output of the sun but obviously we can't do that right now so I have to like wait a couple decades. Okay, my mistake. Well, at least for the next few decades I guess.

Anyway, the total employment up and down the supply chain for these buildouts is very hard to precisely calculate but it is in the hundreds of thousands according to most industry estimates. A lot of this construction is also taking place in remote rural areas where these companies have been able to lobby for favorable local tax and regulatory treatments. Now, that whole game is a separate problem that I know that the old team over at Micro is actually making a video on. So, I am not going to steal their thunder on exactly how messed up this can all get. But the point is, as it relates to employment, a lot of on-site workers doing this construction are only in these locations while buildouts are happening. That's very stimulating to the local economy for the two years it takes to get these facilities online. But if it slows down nationwide, it could have broader employment impacts beyond just the workers directly in the supply chain.

Now, in the short term, even if this technology never becomes commercially viable, it could be written off as tech companies using the piles of cash they have accumulated to fund their own jobs program for a couple of years. It would be like if the wealthy old man from Up the Road paid you and your friends a million dollars to dig a giant hole in his backyard. Would there have been a better way to use those resources? Probably, but it wouldn't cause any other ongoing problems, right? Well, it might actually, and we will get to that. But right now, the concern is about the existing problems it is covering up.

$400 billion in capital expenditure this year alone is a pretty big economic stimulus. Nationwide, according to data from Bloomberg and Renaissance Macro, company spending on AI is now contributing more to GDP growth than consumer spending. That money is ultimately flowing from the big bank accounts of big tech firms and investment funds into the hands of thousands of businesses, suppliers, and workers. In fact, if we just look at the numbers alone, then in terms of money being pumped into an otherwise unhealthy economy, this is almost exactly the same as the combined stimulus and troubled asset relief program spending from 2008 to 2009 to stabilize the economy after the GFC.

Now, of course, these payments are not as direct. And for now, the good news is that most of this money is still coming from cash reserves these companies already had with borrowing mostly being used for cash flow purposes. But it should still give you a sense of scale as to how much economic activity is being generated on the contingency that AI hype continues. If that gets taken away very suddenly, it could quickly reveal how shaky everything else was under the surface.

Then of course, there is the wealth effect of people watching the green line go up. A lot of retirees and independently wealthy households have seen their investment portfolios almost double over just the last 5 years, which means that they have been a lot happier to spend conspicuously. This spending may be very uneven, but in theory at least, someone renovating their vacation home or buying a fifth sports car is at least giving money to contractors or a commission to a car salesman.

A lot of people living off their investments actually have a set draw down rate of about 3 to 4% for their everyday spending. You might have heard about the 4% rule for early retirement. And this rule prescribes that you don't spend more than 4% of the value of your invested assets every year so that you don't eat into your principal wealth. Now, if your portfolio doubles in value over 5 years, that means you can do twice as much spending without changing your 4% rule. Now, a lot of wealthier households are not this mathematically rigid. But the wealth effect happens subconsciously as well. My good friend Ben Felix has done a series of great videos on this subject, as well as a video he made last week analyzing the current risks in our very concentrated investment market. So, if you're interested in a detailed breakdown of these ideas, I will leave a link to them below.

If people see the value of their homes double, even though it's difficult to actually access those funds, they are still going to be more likely to get themselves to do something like a big remodel. Now, so far, this has actually helped to keep things ticking along. But it's also no secret that a lot of these companies might be a teensy bit overvalued. The only market worth disrupting after burning this much capital is the labor market. Even a small slice out of the $12 trillion worth of annual payroll could recoup these costs pretty quickly. But without that, investors are only going to have so much patience.

Even then, it's not exactly this straightforward. One of the biggest ironies in this push to automate jobs is that the tech bros are trying to automate jobs that barely make any money in the first place. Uber drivers subsidize the wear, tear, and depreciation of their cars, and the taxpayer frequently subsidizes warehouse workers with food stamps. The financial reality behind taking these kinds of jobs with expensive technology is questionable at best. And if too many people start asking those questions, that could be a problem. If these values are reconsidered in a major way, then both infrastructure spending and investment fueled household spending would fall simultaneously, seriously hurting employment. We have always wanted Green Line to go up, but never before has so much real employment depended on it.

If this technology does turn out to be everything investors are hoping for, it's going to take our jobs. And if it becomes clear it can't do that, it will very likely crash a very frothy and concentrated market and we are going to lose our jobs. It sounds pretty bad, right? Well, there is actually a third option. It's almost become the assumption that this will all end spectacularly, but there isn't actually any guarantee of that.

For now, the companies involved have more than enough cash between them to cover outside debt. And even if less liquid players like OpenAI can't meet the record-setting spending obligations with companies like Oracle, the recourse to actually collect on those payments isn't very strong. Nobody wants to sue the next person along from them in the financial circle. >> I'm so happy that my lawyers have something to do. I'm so happy that I get to sue how money works and show them how the legal system works. >> Well, almost nobody, I suppose.

Anyway, to keep their own market stable, the most likely thing to happen is that these companies just take equity in exchange for compute time or quietly renegotiate commitments behind the scenes and just stretch them out over longer periods of time. So, even if advancements continue to slow down, this could plot along for a lot longer than you might expect. But in the long term, this kind of scenario could actually be an equally damaging outcome. Consistent investment into AI above anything else is going to starve other hopeful technologies of getting the funding they need for development. The joke about putting AI in the name of the company to get investment is true, but a lot of companies do it because they need that investment to develop their product.

An anecdotal example I saw while putting this video together was the startup Boom Supersonic. It has been raising money for the past decade to develop a modern supersonic passenger jet aircraft. Now, maybe you think faster air travel is a worthy technology to pursue. Maybe you don't. But that doesn't really matter. The point is that this month they came out with a new product which was basically taking one of their jet engines and attaching it to a generator which they were pitching as a new and innovative way to power data centers. Now, this of course is just one example, but startups trying to develop new technologies are all being forced to play some version of the same game to keep funding going.

On a larger scale, there has been a big push towards reshoring manufacturing. This may very well be a noble goal with strategic value, but it's also made much harder by rising energy costs and electrical grids that need to support factories as well as data centers. Right now, the data centers can just afford to pay more, which means factories and regular households have to suffer the financial consequences. The same kind of financial crowding out is happening in a lot of markets already. The reason you are going to need to sell a kidney to afford 2 gigs of RAM is because the entire production base is being redirected towards serving this one industry. Going all-in on one hand may or may not pay off, but it also means we don't have any chips left to take better bets if and when they come along.

And while most people are focused on the obvious players in this bet, one very old, very quiet company has somehow ended up holding more chips than almost anyone else. Larry Ellison briefly became the richest man in the world after his company Oracle delivered worse than expected financial results. As a clear sign of a very healthy and totally normal market, the tech company reported lower earnings per share in revenue than it projected. But despite this, its value soared by over 30% in a single day, gaining more market cap than the entirety of McDonald's within 8 hours. This is all from a company that has historically provided boring software subscriptions to boring enterprise customers. And if an industry reputation is to be believed, they didn't even do that particularly well.

So, as you might have already guessed, this recent boom has had a lot to do with big promises about the future of AI, which is nothing that special by itself anymore. But the way that Oracle is capitalizing on this hype is unique, and it's also worth understanding because unfortunately, it will impact you. Oracle is going to be effectively taking over control of TikTok in America. It has massive influence in both Silicon Valley and Washington. It might single-handedly be keeping the AI bubble inflated. Oh, and this is not to mention the direct connection it will have with potentially the biggest traditional media company of all time. It's all very impressive stuff when you remember that unless you were unlucky enough to have to deal with their software services, most people have no idea what this business actually does.

So, there are really three things you need to understand about this company. The financial shenanigans it has pulled to suddenly explode in value. The way it's using that paper value to grab real value and how it's using that value to become one of the most powerful institutions in the world. I know organizations like BlackRock usually get all of the attention for pulling strings behind the scenes. And honestly, some of these fears are not totally unwarranted, but hold on to your tinfoil hats because there is an argument to be made that what Oracle is turning into is going to be much worse. And here's why.

The recent run-up in Oracle stock price has been largely thanks to its aggressive shift into AI infrastructure support. The company is building massive AI data centers and then selling access to them to companies that don't have the money or expertise to build their own. Now, cloud computing is not a new offering from Oracle. And a lot of other companies offer similar products, but the rate at which they are expanding this part of their business supposedly is what has surprised a lot of investors. As far as tech companies go, Oracle is pretty ancient at almost 50 years old now. Investors were treating it like a mature, stable, low growth industry incumbent that had carved out a nice little niche for itself in commercial software services. But over the last year, it has effectively rebranded itself in the markets as a born-again AI startup, complete with the investor exuberance that comes with it.

This really came to a head in their most recent earnings when they announced that they had almost half a trillion dollars in backlogged orders from AI companies desperate to utilize its expanding infrastructure. It's basically that number alone that has fueled the surge in Oracle stock price because everything else in the financials were underwhelming at best. But that didn't matter because big numbers go up and people are hoping this could be the next Nvidia for everybody who missed out on Nvidia. But there are three big problems with that absolutely massive number.

The first is the simple fact that we need to get out of the way first, which is that this spending is probably almost certainly unsustainable. Generative AI does have some useful applications. And in the long run, we are going to figure out more ways to integrate it into everyday technologies. But $455 billion is more than 2% of US GDP. And that's in just one set of contracts from one company. When you are spending that much money, handy little technologies are not good enough. It genuinely needs to radically reshape the world in one way or another. Even if it's not for the best, at least the analysts will be able to see where the money is going. The big players are getting more and more desperate to prove that the scale of disruption is possible while everybody else is getting more and more skeptical. This was covered in a video last month. So, I don't want to retread too much ground here because the economy destabilizing scale of AI investments is not even the biggest problem with these Oracle numbers.

The bigger problem is that later on in that same financial letter, the CEO notes that basically all of this money is coming from just four contracts with three individual clients. So, even if those contracts were split evenly, that means these clients will be spending more than $150 billion each on effectively renting a data center. The logical question is why would companies with that much money not just build their own data centers? Well, that's the third problem. The biggest new client juicing these numbers is OpenAI who has announced plans to spend over $300 billion with Oracle over the next 5 years. The problem is OpenAI doesn't have $300 billion to even have a chance of being able to fulfill this commitment. They are going to need a lot more investor money. Fortunately, and completely coincidentally, at around the same time last week, Nvidia announced that it will be investing a hundred billion dollars into OpenAI. This cash will be used to help develop their compute infrastructure through deals like the one they have made with Oracle. But Oracle is building its compute infrastructure by buying billions of dollars worth of Nvidia GPUs. So uh yeah, all they are doing is effectively turning outside investor money into deferred revenue on the next company over which is making these businesses look more promising than they really are to even more outside investors.

Now AI has been playing a game of say a bigger number for a while now to the point where last week Sam Altman unironically suggested that we could build a Dyson sphere to power core compute capabilities. Now, this is all really dumb, but it should only be a risk to the people investing money into these businesses, and they do need to do their own due diligence. The real problem for everybody else is what they are doing with these fantasy valuations once they have achieved them. So, it's time to learn how money works to find out what Oracle and the big man Larry Ellison are really up to.

Oracle and its founder, Larry Ellison, have been extremely wealthy for a very long time now. And over the last four decades, neither of them have been afraid of throwing their money around a bit. Oracle spends millions of dollars every year on political donations and lobbying, and Ellison himself was one of the biggest spenders on last year's election, even if he was a little bit more low-key about it than Elon Musk. Senior executives in the company, including its outgoing CEO, have also held key positions within the current government. But like all good philanthropy, the alleged return on investment from these donations has been staggering.

On a totally unrelated note, TikTok's parent company ByteDance must sell off its US operations in a deal that has been heavily monitored by the government because of the sensitive user information and influence the platform controls. To facilitate this transfer, Oracle has been selected and approved as the company that will handle the new platform using its existing cloud infrastructure. This is going to give it influence over the algorithm and access to that same sensitive user data. Now, you might reasonably say that it's better for an American tech giant to have this data when the alternative is a Chinese tech giant, and that's completely fair. What hasn't been completely fair is the tender process to decide which particular tech giant would be gifted this valuable opportunity. The details of this acquisition are still incredibly vague and opaque, but almost every independent estimate says that the $14 billion valuation put on the social media platform is well below what it should trade at. Now only the American operation is being forced to sell here. But even still, comparing it to a competitor like Instagram based on sales within this market alone, it should conservatively be worth 10 times more.

Now, political lobbying potentially and totally coincidentally resulting in a lucrative government deal might not come as a huge surprise to you anymore. But well, >> say the line, Bart, >> it gets worse. The Ellison family have used their wealth and influence to move far beyond one single tech company. Now, I have to give a warning that this next part involves looking at the world's most punchable face. Although nothing I can say can truly prepare you.

Larry Ellison's son, David Ellison, started the company Skydance Media in 2010, reportedly with seed capital of $350 million coming from his father and other associates. Last year, Skydance took over Paramount to form Paramount Skydance, which Ellison still has voting control over. This new mega media company is also reportedly eyeing a deal to acquire Warner Brothers, which would make it one of, if not the largest traditional media conglomerate in history. This is incredibly impressive from a business started just 15 years ago. Although funding from one of the richest men in the world has certainly made rapid expansion a lot easier, but the cozy and at times lucrative relationship with the government has raised some eyebrows around how this business is run, such as the cancellation of Steven Colbert at the same time that the TikTok deal and the current slate of big acquisitions were being assessed. However, the company insists that it was purely down to the show receiving poor ratings. Now, that may very well be true. Late night talk shows have been a slowly fading medium for a long time now. However, even the potential for an organization with this much influence to have this much control is concerning enough by itself. But it gets worse. As something of a synergistic move with Oracle's heavy focus on AI, David has been a vocal supporter of using the technology to lower costs in the movie production industry. So basically the worst use case for AI.

This is also all ignoring the biggest problem of all. Larry Ellison is 81 years old and while he looks pretty good for his age, he is eventually going to die. The exact details of his estate are unknown, but it's safe to assume that he will pass down at least some of his influence over Oracle to his son. That means one man could soon potentially control one of the biggest social media platforms and traditional media platforms at the same time. So if you ever wondered what happened if we mixed Rupert Murdoch with Mark Zuckerberg, we may be about to find out.

But Oracle and Larry Ellison are really just one of the most extreme examples of a much broader trend. The most powerful position in basically any industry right now isn't making the product or selling it. It's owning the middle. The biggest companies that you never heard of make billions of dollars every year by inserting themselves as wholesalers, distributors, licensers, or aggregators, getting between you and the factories that make your stuff. The only problem is everybody kind of hates middlemen, which is why entire industries have been created from the ground up to cut them out by offering direct to consumer, peer-to-peer, direct selling, disruptive, streamlined outlet platform solutions to make consuming everything an easier, faster, and cheaper experience. The only problem is it never really works. And in our venture capital fueled rush to cut out the middleman, we have just created even bigger ones.

>> I simply put, we cut out the middleman. So we design our own frames. We work directly with manufacturers. >> Yeah, we cut out the middleman and voila. >> Uber on Monday said it's cutting 3,000 more jobs in a second wave of layoffs. >> More and more Americans are cutting out the middleman in the prescription drug industry. >> We cut out all of the middlemen. >> No wholesalers. No middlemen, no distributors, no importers, just direct to your front door.

So last month, the online coupon code extension Honey was exposed for basically being a massive fraud that would claim referral revenue from its users online shopping, even if they found no coupons to offer, and even if shoppers were directed to the website from another affiliate that actually did most of the work of selling them that product. They inserted themselves as a useless middleman and allegedly claimed billions of dollars in referral revenue. Now, unless you are living under a rock, you have probably already heard all about this. The original expose video has over 15 million views, and basically every major influencer on the platform has come out to call it the biggest scam in YouTube history. It's a big claim considering large YouTube sponsors have given unlicensed mental health advice, leaked people's personal data, frozen their life savings, or misappropriated their investments. Hot take alert. But a multi-millionaire like Lionus not getting his referral revenue from Amazon is simply not as big of a deal as someone losing their life savings on a bank they were told to sign up for by a trusted financu.

Now this is not to defend Honey in any capacity. What they did was clearly highly unethical at best and will probably be proved fraudulent in the many upcoming court battles headed their way. But you would be forgiven for thinking that the reason it's getting so much attention from YouTubers is because this sponsor was stealing from the YouTubers themselves. The art of good business is being a good middleman. All of the biggest channels on this platform are made possible because at some point they may be able to convince you to buy something. That means they can directly or indirectly be rewarded for that through advertising. The reason this has rustled so many jimmies is because someone beat them at their own game, even if they weren't playing fair.

Now, in the interest of total transparency, this channel is just as reliant on advertising to pay our team as the other big creators. And while we don't take sponsorships from financial firms because of the potential for you to lose your entire life savings, we honestly probably would have worked with Honey if we were offered because just like anybody else, we didn't see anything wrong with them. So, all right, with that out of the way, this video is not about the Honey scam in particular. It's about what it represents.

If you buy something like this stylish personal neck fan off Amazon, Amazon itself doesn't actually sell the product. It just provides a marketplace and distribution services to a third-party distributor. That third party distributor buys these products from an Alibaba wholesaler in China who themselves work with white label suppliers who work with manufacturers. Even for a relatively simple piece of consumer junk, there were five middlemen between you and the person actually making that stuff. If you include an affiliate link in there, there's six different businesses that need to be paid, cutting into the profits or more likely raising the price of what could be a simple exchange. Consumer retail isn't even the worst example. Consultants, brokers, and fund managers, and technology platforms have all built some of the most lucrative businesses in history by capitalizing on the golden age of the middleman. But why is this happening now? Middlemen are nothing new, but there are a few reasons why it's become particularly lucrative in just the last two decades.

The first reason is we have simply mandated middlemen. The list of corporate regulations only grows longer every year and so too do industry guidelines, best practices, frameworks, reporting standards, certifications, privacy controls, and self-imposed goals like ESG, DEI, CSR, GE, GRC, and whatever other BS McKenzie cooks up in the new year. Some of these rules are absolutely necessary and genuinely help both consumers and workers. Some of them are just a little bit silly and some of them only exist to make sure that a group can make money by consulting on the issues they create. Looking at you car dealerships. But either way, there are a lot of rules that businesses have to adhere to. Out of this, it's become very lucrative to provide third-party solutions that offload compliance onto another company that is better handled to deal with it. This both protects businesses from legal repercussions and lets them focus on where they actually add value in their business.

Visa and Mastercard are some of the most valuable companies in the world and their business operates exclusively on middlemaning transactions between credit card processors like Stripe or Square and banks who are themselves middleman between the customer and the business that they are buying from. Businesses could just take cash or work out their own payment system, but it's just way easier to pay a few cents to Visa to process the transaction and let them handle the accounting, computing, and dispute resolution. It's also not like businesses have much of an option. When was the last time you paid in cash for anything? Lock your answer in the comments below. Of course, a small group of people live and die by the Benjamins, but the vast majority of consumer spending these days is done with credit cards, debit cards, or more recently, buy now pay later services. Beyond letting people spend money they don't have, they are just easier. Any business that didn't want to work with these middlemen would immediately make it very difficult for most of their potential customers to shop with them.

Now, this isn't a problem by itself. Modern payment providers have made doing business a lot easier for everybody, and they should be rewarded for providing that service. But as things like this get more complicated, every extra step of the supply, payment, or marketing process is an opportunity for a business to gain market power and squeeze out a little bit more from consumers. Even benign payment providers with lots of competition are causing a long list of problems for consumers. But that's got nothing on what the middlemen are doing to regular workers.

To kick off the new year, a lot of people have been talking about H1B skilled worker visas. These are something we have talked about a lot on this channel before because they give companies the opportunity to employ workers from overseas if they can't find any workers in America willing to do the same job. Those foreign workers are then dependent on their employer to maintain their residency, which makes them great employees because they really can't quit unless they find another company willing to sponsor them. A lot of companies, especially tech companies, clearly prefer these workers because they have more staff loyalty and won't argue too hard for a pay increase. The rules and restrictions on these visas are pretty extensive. And the rules are supposed to stop companies undercutting American workers and just hiring more subservient expat workers instead. But it's really easy for big companies to get around. If you are the CEO of a tech company and don't want to pay tech developer money for your engineers, just put up a job posting paying well below market rates. When nobody wants the job, you can then claim that you need to access someone with distinguished merit since nobody in America can fill your role. It's a broken system that's not great for American workers who get undercut for jobs, expats who get exploited by the companies they rely on, and genuine immigrants trying to earn their way into the country. But it's great for your bottom line.

The problem for you is that companies are only allowed to hire so many H-1B workers. So, a new industry has developed where large consulting firms will hire as many H-1B workers as they can. Tech companies like Google, Meta, and X are getting a lot of attention for their exploitation of these programs. But according to the Department of Immigration records compiled by the New York Times, the largest H-1B sponsors have actually been middleman firms that bill out their consultants hourly. This report was initially published over a decade ago, but more recent data still shows that distinguished labor hire agencies have learned to game the system. This lets companies offload the legal and reputational risk of using these workers onto the middleman consulting firms. And even with a small premium for their services, it's still cheaper than paying market rates for technical skills.

But the regulations arms race is only the first reason why we have entered the golden age of the middleman. A long time ago in an economy far far away, the closest thing we had to businesses producing products for consumption were smiths, saddlers, coddlers, and other ye old tradesmen. These workers made things by hand only after an order was placed with them. Given their basic tools and lack of production lines, they didn't have an abundance of goods that they had to worry about finding a buyer for. They just made enough to keep up with orders. This really changed with the industrial revolution and true mass production where consumer goods could be made faster than people could buy them. Over time, the modern business model of a manufacturer selling to a wholesaler who sold to a retailer who sold to a customer started to emerge because it worked.

Before this turns into a how history works video, every step of this process has a part to play in making sure commerce is conducted smoothly. There have been hundreds of startup companies that have sold themselves on using a direct to consumer model to cut out the middleman. Even the biggest direct to consumer startups with the most investor money like Casper mattresses, Bonobos Tailored Clothing, Movement Watches, and Everlane Fashion have either faced significant financial difficulties, been acquired by a traditional retailer, or gone out of business entirely. The YouTube channel Modern MBA did a great deep dive on Casper mattresses in particular, analytically pulling apart their broken business model. If you're interested in thorough business analytics, I'll leave a link to his channel below. The point is, effective middlemen don't actually make businesses less efficient. If anything, they actually make the whole process too efficient. The manufacturer's job is to make stuff as efficiently as possible. The wholesaler buys from the manufacturer in advance to make sure that they have consistent cash flows, and the retailer handles customer service and sales, but relies on the wholesaler to make sure goods are available to sell. The wholesaler may look like a useless middleman, but they are a buffer to make sure that manufacturers don't have to worry about who they are going to sell their junk to and retailers don't have to worry about not having enough junk to sell. That's the idea anyway.

Some companies are big enough that they can be both the retailer and the wholesaler, which is why Walmart and other big box stores will be able to outcompete local independent retailers. Now, eventually these companies figured out that it's cheaper to do the manufacturing offshore, which meant supply chains got longer and more complicated, making the job of the wholesaler even more important and potentially more profitable. Today, the only retailers that can really compete on value are major corporations that integrate wholesale into their operations or get around it all together. This has given them more market power over both producers and consumers because the barriers to entry into these markets has become so much steeper.

We aren't living in the golden age of middlemen because there are more of them. In fact, it's actually because there are less of them. They just have a lot more power. Big companies have realized it's a lot easier to gain market power as a middleman than as someone who deals with the end consumer. Katherine Judge is a law professor at Columbia and published the book Direct: The Rise of the Middleman Economy and the Revolution Underway. In her book, she outlines how technology, globalization, and financialization have all worked hand-in-hand to allow a small group of companies to become middlemen in business operations that just weren't possible that long ago. In particular, she focuses on the collection of data as a tool that is uniquely useful to middlemen because being right in the middle of operations means they can access information from both suppliers and consumers. Effectively, leveraging data is an incredibly powerful tool to reduce costs on one side of the business and maximize sales on the other.

Even if you ran a business that could somehow supply products with the same economies of scale as Amazon and Walmart, you would still find it incredibly difficult to compete with them because they have so much data about when to order stock, where to build retail stores and distribution centers, how many drivers need to work a particular route, when people are most likely to make a purchase, and what items should be presented to them first as they log in or walk into a location. It's all data, and being the biggest middleman in their respective fields allows them to collect more of it and use it in more ways. Other companies can purchase data from data brokers or even use select meta analytics provided by the big tech companies, but it's never going to be as comprehensive, detailed, or as deep as the good stuff they keep for themselves.

The final point that Judge makes is lobbying power. These companies can argue for more or less regulation in the fields that are important to them extremely effectively. Votes over boring middleman regulations barely get lawmakers to show up, let alone get the interest of the general public. So, this is amongst the most effective ways for these companies to lobby. Now, the solution to this problem is the same solution to a lot of problems in corporate America. Clamp down hard on companies with too much market power, especially those that are not as obvious as the Googles and Facebooks of the world. But I wouldn't hold my breath. In the meantime, go and watch this extended cut video next to find out why when everything is a crisis, nothing is. And don't forget to like and subscribe to keep on learning how money works.