This is what everyone's building right now in AI. But most of these AI startups, they won't last more than 18 months. So where is the real money being made?
Let's dive into it and learn how you can become rich without writing one line of code. Section one, the six tiers of AI. Here's your 5-minute MBA in AI economics.
[music] Tier zero, energy infrastructure. AI lives in the cloud, but it feeds off electricity, which is why data centers are projected to consume more energy than entire countries by 2030. And if you're thinking like a person who likes money, you're going to be asking yourself, who's powering the power?
We're looking at utilities, generators, grid plays, or even buying or starting businesses that service energy infrastructure because look at this consumption level. Actually, wild companies like Hanley [music] Energy, which builds data centers. The play, if you're not going to build the data center, come the specialist, like the contractor for emergency 24/7 data center services in a specific region.
So if I was you, what would I do? I would copy the homework of these guys, Hanley [music] Energy Electric. So they're smaller contractors.
They build expertise around data [music] centers in particular. And if I was you, I might be looking to become a specialist, a contractor for them, knowledgeable about the area. And I really think this next tier you're going to see chips, this is the arms race.
[music] Nvidia, AMD, and whoever controls chip manufacturing now controls global innovation. [music] But I'm not just interested in the stock. I want to know who's building the fabs, who's maintaining the clean roofs, who's doing the HVAC and installation.
I think most people think, [music] "Hey, this is just for the big guys. I'm out. " You know what?
I guess you're right. But the contractors building [music] these 10 billion dollar facilities, they've been around for 20 years. They have certifications and engineering teams [music] that speak in acronyms.
We can just be the guys who build the things those guys need in the real world. And I think those billiondollar fabs, guess what? They also still get dirty.
They need HEPA [music] filters that clog. And the big contractors, they don't want to send their $200 an hour engineers to change filters. That's where we come in.
And we just charge them $150 an hour. These are the specialized service [music] companies nobody's talking about. Talk about like Promera, formerly DataClean.
They started cleaning these data centers and clean rooms. [music] Now they're the go-to across the entire industry. So like, do one thing obsessively well.
Keep mission critical environments particle-free. You don't have to know a ton. You just have to show up at the shop back and say, "I clean good.
" Then we've got tier two. Tier two is the data centers. $455 billion went into data centers last year.
That's up 51%. That's like so much money I can't even quite understand it. Microsoft, Meta, Amazon, all the boys spending billions here.
What about the plumbing, roofing, wiring, insulation, the local businesses that these AI empires can't function without? This is AI translated for bluecollar domination. And look at how much money these companies are making today.
I think this is the kind of unsexy business that makes people rich while everyone else chases AI unicorns. So JM Tech Group, they clean data centers. Thrilling, right?
No, they're not just cleaning. They're actually looking at fancy detection systems that are missed, fire hazards, disconnected cables. The play here is really simple.
You show up to clean. You show up to inspect. You show up to fix.
Suddenly, you're not the janitor. You're the riskmanagement consultant. And guess who gets paid a lot more?
the riskmanagement consultant. I think people in this space are going to make an ungodly amount of money. This is the picks and shovels of the AI industry and nobody is talking about it.
Tier three of AI, we've got foundation models. This is where the big dogs play. Open AAI, X, Google, Anthropic.
These are like the oil rigs. Big money, long timelines, uncertain returns. You and I aren't going to invest there.
Most people watching, we can't compete. But you don't need to build the oil rig. You just need to sell the fuel tools or whatever tacos to the guys on it.
I agree with what another investor, Elizabeth Yin, said. The typical early AI startup, they are burning, aka spending, $500,000 a month. A lot of this is on GPUs.
If you don't know what that is, it's basically like a super fast calculator built to do thousands of tiny tasks [music] at the same time. You need that in order to power your AI. At this pace, if you even had $10 million as an AI startup, it only buys you 20 [music] months of runway, aka ability to live and still survive as a company.
But during that period, that's no profit because the problem is you have too many users and each one costs you money. So, OpenAI hit a $300 billion valuation with revenue growing from 3. 7 billion to 12.
7 billion. But the caveat here is that the training [music] costs for these companies, they grow exponentially. And so, future models are going to cost them billions of dollars to train.
Think of this like the plumbing layer of AI orchestration tools, APIs, deployments, frameworks. It's like the AWS, the cloud of AI. It's not sexy, but neither is Stripe or Data Dog or MongoDB.
And yet, look at how much money those companies are worth. Multi-billion dollar businesses cuz they make everything else work. I think some people are worried that bad things will happen here.
That the big tech and AI companies are all well sleeping with each other. Take a look at this chart from Bloomberg. I think it means they actually are all sleeping together.
You know, these companies are servicing one another. Nvidia and AMD are investing in companies they sell to. Anthropic is using Amazon's web services and Amazon is an investor and Anthropic and Microsoft is invested in OpenAI and counts OpenAI as a customer.
This is actually called circular financing [music] where dollars trade between firms clouding actual demand. So are they making all that revenue or are they just [music] paying each other? Then we've got tier five, the AI native applications.
This is like the shiny stuff. Most people playing here are building without a business model. But if you focus on apps that replaced cost centers and drive real productivity, you got a shot.
That's where we're hunting. Now the question becomes, how do you not go alone here? Tools like Replet and Cursor will let you play Startup Founder for a weekend.
You'll build something that looks legit, feel like a genius, maybe even get a few users. Cool story. But when it's time to actually scale, when the demo breaks at 2 a.
m. and customers are screaming, when you need to connect payment processors, databases, APIs, and all the back-end plumbing that makes real businesses work, you're how do the kids say, cooked. So, if you're going to play in this space on this tier, here's the kicker nobody talks about.
If you actually win in this space, big tech becomes your biggest threat. Not because they'll copy you. They might, cuz they'll steal your team.
If you build to millions, tens of millions, maybe even unicorn status, and suddenly Meta slides into your senior engineers DMs with a $2 million package and unlimited Lacroy. Who can say no to that? You know, this is called key person risk on steroids.
So, if you want a real example, Google Meta and Windsurf. It's like this nerd soap opera from Silicon Valley that'll make you want to lock your engineers in the basement [music] with golden handcuffs and NDAs written in blood. Don't do that though.
That's not That's not good. So, the play here is if you're serious about tier five, you need a technical co-founder with equity, not just salary. You got to have something called vesting schedules that actually mean something.
Need retention bonuses that kick in [music] before the big guys come calling and a culture so strong your your team turns down Zuckerberg and everything he can offer. or you accept that you're building a nice little one to five million dollar lifestyle business which I think is awesome. Either way, know what game you're playing because in AI apps, your engineers aren't just employees, they're the entire moat and Silicon Valley knows it.
Now, the scary part in some ways is as the Financial Times put it, America is now one big bet on AI. AI better deliver for the US or its economy and markets will [music] lose the one leg they are standing on. Now, look at this chart.
It's kind of horrifying. So, the next question is, is this a bubble? And the answer is like probably 400 billion being spent on AI infrastructure.
Some AI companies raised $2 billion. The S&P almost 40% of its gains this year are tied to 10 AI heavy stocks. So if it looks like a duck, if it talks like a duck and it quacks like a duck, it's probably a bubble.
But here's what makes this one different. Unlike the tulip craze or even parts of the. com bust, [music] AI is already in the economy.
Like this is kind of crazy. 90% of GDP growth, which is basically a fancy way of saying like all the money the economy makes AI related. The biggest companies are trading at kind of reasonable multiples compared to the big bubble back in 99.
So what do you do? I'm not telling you to go buy tech stocks. [music] In fact, I don't think you should.
You don't short it. You don't blindly ape in either. You build a moat because you don't actually need to code.
You don't need VC. You just need cash flow. Which brings us to the fun part.
One of the fastest way to build cash flow in the digital economy is launching a digital storefront. And I've been experimenting with a new tool called build your store. And it's basically like having an AI store builder.
Here's what it does. [music] You tell by what niche you want to target and let's say pet grooming or home office upgrades. [music] The AI instantly builds your entire Shopify store, products, design, everything.
You get a ready to sell business in minutes with zero tech skills and [music] zero inventory to start. And as a part of this setup, Build Your Store also walks you through a quick AutoDS registration, which basically what it does is automatically add 10 products to your store and allow you to automate shipping and order fulfillment. And this AutoDS 30-day trial is just a buck, $1.
[music] And the best part, ProLevel Design you normally pay 200 bucks for already included. So, if you're looking for an online side hustle with lower risk and potential for lots of upside, this could be one of the funnest plays I've seen. So, thanks to Build Your Store for sponsoring this [music] video.
Head to the link in the description to your Shopify store with build your store AI today. And now section three, how to profit either way. So whether this thing pops or rockets, here's where the money's made.
One, you could leverage AI for margin expansion. This is a fancy way of saying make more money [music] in your business. AI is a tool, not a business.
So, in our business right now, and in the 14,000 people we teach, how to build more profitable businesses at Contrarian [music] Growth Boardroom, we use agents to automate customer service and admin, pre-qualify leads, [music] build SOPs faster, cuts costs. Like, for example, this furniture company, Temple and Webster. They didn't build a new GPT app.
They simply layered AI across support, [music] product descriptions, shipping calculations, customer service. So, 80% of interactions are now automated. The result was costs are down 60%.
That means that they're making 60% more money. Two, buy boring serve. Open AI, Anthropic, XAI, Google, all racing to make chat box that each sound a little smarter than the last.
When every product looks and acts the same, what happens? Prices drop. Amazing for users, tough for companies.
So, with open- source and Chinese models giving kind of this commoditized market, it's not good. And what I think Goldman says perfectly is competition is often underestimated and the returns on capital invested overstated. fancy way of saying real winners in the coming years won't be model makers but scrappy operators.
So I want to buy businesses that wire the data servers, cool down server [music] farms, clean semiconductor labs, install solar. These are just the businesses behind the boom. Services like these are going to be killing it.
Three, don't be an idea guy. Be a workflow fixer. You want to build?
Great. But solve real problems first. The money won't go to the flashiest chatbot.
it would go to the one that saves law firms 40 hours of doc review like co-consil legal. It plugged into things that were already happening like document review, contract drafting, and it cut the time to do that work in half. So, an AI strategy like this will make you twice as much money.
That's [music] your sales pitch. Four, stop waiting, start owning. By the time regular people feel safe, the money's already been made, you guys.
You got to use AI to speed up ownership now because it doesn't eliminate opportunity, it actually multiplies it if you move fast enough. I think one of the biggest opportunities in this space is using AI before everybody else [music] does. People take their time to figure out.
An underrated bottleneck in AI isn't computing power, but human creativity. Companies can sell you endless new models, but if you don't really know how to wield them, then what good are they to you? The real opportunity, learn how to apply these new tools in [music] ways that actually make you more money.
Harvard and Stanford found out that some workers are spending more time correcting AI output than doing actual work. McKenzie found twothirds of respondents say their organizations have not even started using AI. [music] So if you can get ahead of these trends, you're just going to make more money.
The play, the biggest benefits will likely come in the form of AI agents. This is the last one. These are the tools to help you.
Many of them are okay today, but digital specialists will change [music] everything. Agentic AI understands goals and can actually plan for you. They can make decisions, take initiative without constant human direction.
This is all from Vista Equity, what one of the smartest software companies in the world. So for you, that will eventually mean [music] having a little team of workers for you in AI. And I think one of the reasons that's so powerful is the key to making money in the AI age is not letting AI think for you, but making it think with you.
Most people let AI copy what's already out there. >> You be original. You move fast.
AI can only generate, but you can direct. So I don't actually think you should try to time the market. It's time in the market that's more important over the long term.
Here's the thing. Nobody, not us, not Sam Alman, not Elon Musk knows exactly what's going to happen with AI. So, you don't have to predict it.
But what if you built the picks and shovels? The way you win is by owning the [music] things nobody else thinks is sexy. So, bubble or boom, that's how you stay rich.
You learn how to use the tool instead of being used by it. And if you want to learn exactly how to do this, I'm teaching a [music] free workshop on how to use AI to buy cash flowing businesses. So, let's do this [music] together.
Let's be part of the picks and shovels that last way beyond the gold rush. I'll see you there.