Meet the AI Cash Machines: 10 Stocks Turning AI Into Real-World Profits
Listen to the audio version of this article (generated by AI).
In the 3,000-year history of the insurance industry, nothing had ever moved this fast.
That’s the record AI-powered insurance company Lemonade says it set on December 23, 2016.
At 5:47pm in New York a customer named Brandon Pham hit “Submit” on a claim for a stolen Canada Goose Langford Parka worth $979.
Three seconds later, the claim was settled.
AI Jim — Lemonade’s claims bot — reviewed the claim, cross-referenced it against Pham’s policy, ran 18 anti-fraud algorithms, approved it, wired the money to his bank, and closed the case.
Human agents, adjusters, and clerks used to do this work at a fraction of the speed. Today, Lemonade settles about 40% of claims with no human in the loop.
Lemonade isn’t alone. Across a widening slice of the economy, AI is now doing the work behind the scenes.
Right now, for instance, an AI model is approving someone for a loan in about the time it takes to pour a cup of coffee. A company called Upstart uses one to weigh the application — not just a credit score, but the applicant’s income, their job history, the rhythm of money moving in and out of their account — and make the call in seconds.
The company says this allows it to say yes to more people, at better rates, with fewer defaults.
AI is also working around the clock to decide what ad you see next online.
In the half-second it takes a web page to load, it’s already sized you up — what you tend to buy, what you seem about to want — and picked the one ad, out of thousands, most likely to get you to open your wallet.
A company we’ll look at later in the report does this billions of times a day for some of the largest brands on the planet.
Even your favorite restaurant may be leaning on AI now, without you ever noticing — an algorithm deciding when to reorder before the walk-in runs empty, or when to add a shift before Friday night gets slammed.
For most people, “AI” still means a chatbot — something you type a question into and get an answer. But AI is now also doing the unglamorous work behind entire businesses — underwriting, adjusting, targeting, scheduling — jobs that used to take a payroll of people to do.
We call these companies AI Cash Machines — a name that isn’t subtle, and isn’t meant to be.
These are businesses where AI has taken over the actual work of making money: approving the loan, settling the claim, picking the ad, filling the shift.
And because software doesn’t need a bigger office or a bigger payroll to do more of that work, the profits tend to show up faster, and stick around longer, than they would at a company still running the old way.
This shift isn’t leading the news cycle yet. But that’s no surprise. Most folks are still watching the staggering sums flowing into the AI infrastructure buildout.
Multi-Trillion Dollar Boom
Since ChatGPT’s launch in November 2022 kicked off an AI boom on Wall Street, the conversation has mostly centered on the model builders and the infrastructure that makes them possible.
Who makes the fastest chips? Who’s building the biggest data centers? And how many billions Microsoft, Amazon, and Google are spending to pour the concrete and lay the cable of the AI age.
It’s a staggering sum.
Since ChatGPT arrived at the end of 2022, spending on chips, servers, networking gear, and data centers has topped $1 trillion. And it keeps accelerating.
In 2026 alone, four companies — Amazon, Microsoft, Google, and Meta — are on pace to spend roughly $725 billion on AI infrastructure, up 77% from the year before. And that figure is poised to rise again in 2027.
That infrastructure has to be laid down and paid for — you can’t run AI on good intentions. And there’s still money to be made on this trade.
But as we know from history, the real fortunes in technology booms aren’t made by the infrastructure builders themselves. They’re made by the companies that build on top of that infrastructure.
We’ve seen this sequence play out before. More than once.
Lesson From the 1990s
Think about the internet.
Through the late 1990s, an enormous amount of money went into building the infrastructure that made the Web possible — the cables, the servers, the fiber strung across the ocean floor.
That buildout was real, and early in the boom, the companies that built it were some of the hottest stocks on the market.
Cisco Systems, whose routers and switches formed the backbone of the internet, briefly became the most valuable company on Earth in March 2000, with some on Wall Street predicting it would be the first company ever to reach $1 trillion.
That didn’t happen. Cisco’s stock plummeted 89% in the dot-com bust — and it took 25 years, until December 2025, to reclaim that old high.
But the fortunes that followed — the ones that made early investors rich — mostly didn’t go to the companies that laid the cable.
They went to the companies that worked out what to build on top of that layer. Amazon. Netflix. Google. None of them laid a foot of fiber. They didn’t need to. The cable was already in the ground, waiting for someone with a better idea to make money from it.
Now wind the clock back further, because the pattern is older than that.
Who Builds on Top?
On May 10, 1869, in the Utah desert, a golden spike joined two sets of railroad track and, for the first time, you could travel and ship goods clear across America by rail.
It had cost a fortune and the better part of a decade to build. And within three years, a Chicago salesman named Aaron Montgomery Ward noticed something the railroad men had missed. If a train could carry freight to any town in the country, then a store no longer needed to be a place at all. It could be a book.
In 1872, he mailed out his first catalog: a single sheet listing 163 items. A younger rival named Richard Sears copied the idea. Their catalogs, known as the “Wish Book” and the “Consumers’ Bible,” turned up in nearly every farmhouse in America, and grew into the two biggest retailers in the country. Sears held that crown into the 1980s.
Neither man laid a single mile of track. They built the business that ran on top of it.
That’s the pattern, and it is remarkably reliable. Someone spends the fortune to build the network — the rails, the fiber, the data centers. And someone else, building on top of it, walks away with the larger prize.
This time, the network is artificial intelligence. And the companies building on top of it are only getting started.
Which leaves us with an important question: Who builds on top of that infrastructure the way the last generation’s winners built the trillion-dollar companies of their era?
That’s what this report is about.
We’ve identified 10 AI Cash Machines already putting this to work, while everyone else keeps watching the chipmakers and the infrastructure builders.
They turn up in industries that could hardly be more different: insurance, lending, advertising, medicine, education, and hospitality.
Some of the names you’ll know. Most you probably won’t. A few are still small enough that Wall Street has barely glanced their way — which, for an early investor, is usually where the interesting part begins.
Every boom like this sorts companies into two piles: the ones that figured out how to use the new tool, and the ones that got left holding the old way of doing things.
Nobody wants to own the next Kodak, Blockbuster, or Nokia — firms that failed to keep up with a massive technological shift. This report is about finding the other pile: the companies built to profit from this one.
So let’s meet them.
I. The Deciders
Every business, at its core, runs on a series of decisions. Should we lend this person money? Should we pay this claim? Should we accept this customer, and at what price?
For most of history, those decisions required a human being. A loan officer reviewing paperwork. A claims adjuster inspecting a car. An underwriter weighing risk with a mix of rules, spreadsheets, and gut instinct built up over years on the job.
But that’s expensive, it’s slow, and it doesn’t scale — hire more customers, and you eventually have to hire more people to make those same decisions all over again.
The Deciders in this report have broken that link. They’ve handed over these decisions — the yes-or-no call that used to require a trained professional — to specially trained AI models.
The AI doesn’t assist the underwriter or speed up the adjuster’s paperwork. It is the underwriter. It is the adjuster.
That’s a different kind of company entirely, and it’s why each of the three names below can grow its customer base without growing its workforce to match.
Lemonade (LMND) — The AI Insurer
You met Lemonade at the top of this report — the insurer whose AI assistant settled a customer’s claim in three seconds.
That speed wasn’t a stunt built for a press release. Lemonade now settles roughly 40% of all its claims with no human involved at all, across renters, home, pet, car, and life insurance.
The claims bot is only one piece of the machine, though.
A separate AI model prices every policy the moment a new customer signs up. Where a traditional insurer might take days to underwrite a policy — pulling records, running the numbers by hand, waiting on an actuary’s sign-off — Lemonade’s model does it instantly, weighing dozens of individual risk factors the way a human underwriter once would, but without the wait. And it never stops learning. Every claim that comes in becomes new data, feeding back into the model and sharpening how the next policy gets priced.
A third AI at work inside Lemonade decides where to spend the company’s marketing budget.
Instead of a room full of media buyers making educated guesses about which ads to run and where, an algorithm allocates that spending in real time based on what’s actually converting into new customers. It’s one more job, in one more department, handed over to software.
Put those three AI models together — pricing, claims, and marketing — and you start to see why Lemonade’s growth looks so different from a typical insurer’s.
Pet insurance has become the company’s single largest line of business, built almost entirely on this model. Its newer car insurance product, priced and adjusted the same AI-first way, is growing quickly in a market that’s far bigger than renters or pet coverage ever was.
The number that matters most, and the one worth remembering above all the others: Lemonade’s gross profit has grown roughly eightfold since 2021, while its headcount has barely moved.
That’s the entire thesis of this report, distilled into a single company. It’s proof, in black and white, that software — not more people — is what lets a modern insurer scale.
Oscar Health (OSCR) — The AI Health Plan
Oscar sells individual health insurance — the kind you buy yourself on the exchanges, rather than get through an employer.
But it was built from day one as a software company, not as a legacy insurer with an app bolted on afterward. And that distinction runs all the way through the business.
Behind the scenes, Oscar’s AI prices the plans members buy, based on a detailed read of risk that would once have required a team of actuaries poring over spreadsheets.
It also speeds up claims processing, flagging anything unusual for review and helping clear routine cases more quickly.
And it helps steer members toward the right doctor, specialist, or treatment before a small health problem turns into an expensive one — nudging a member to get a checkup, say, before a manageable issue becomes an emergency room visit.
That last piece matters more than it might sound. Health insurers make money when members are healthy and their costs are predictable. An AI that can guide a member toward cheaper, earlier care — rather than expensive, delayed care — is doing the single most valuable job in the health insurance business.
Oscar has also been licensing that same technology platform to other insurers. In other words, it’s not just running its own insurance business on AI — it’s selling the engine underneath that business to competitors who’d rather rent Oscar’s software than build their own from scratch.
The turnaround inside the numbers was dramatic. In 2024, Oscar swung from a meaningful annual loss to its most profitable stretch ever. Management pointed squarely to the software — not a lucky year in the broader healthcare market — as the reason. 2025 wasn’t a great year for healthcare exchange providers, with a lot more claims than usual. But, while Oscar wasn’t profitable, it did continue to build membership: up 21.8% year-over-year.
One thing worth spelling out plainly, because it’s the one part of this story that isn’t about the product: Oscar’s business leans heavily on government subsidies that make its health plans affordable for its customers.
If those subsidies were scaled back, premiums for a meaningful share of Oscar’s members could rise, which would slow the company’s growth. It’s a real swing factor for the stock — the one piece of this story written in Washington rather than in Oscar’s own code.
Pagaya (PGY) — The AI Lender
Pagaya does the same essential job as a company like AI lender Upstart. But it sits at the other side of the table from the borrower.
Here’s how it works. A bank or a consumer lender gets a loan application it would normally turn down — the applicant’s credit score isn’t quite high enough, or their file doesn’t fit the bank’s rulebook.
Instead of rejecting them outright, the lender routes the application to Pagaya. Pagaya’s AI takes a second look, weighing income patterns, spending behavior, and repayment history far beyond what a simple credit score can capture. And it re-underwrites the applications that a traditional lender’s rigid rules would have missed.
The loans that pass Pagaya’s screen get funded by institutional investors — pension funds, asset managers, and similar large pools of capital looking for steady returns.
Pagaya collects a fee for building that bridge: connecting banks that have customers to serve with investors who have capital to put to work, using its AI as the matchmaker in the middle.
The effect is that Pagaya operates as an almost invisible layer sitting behind some of the biggest names in consumer lending – Ally, OneMain, Klarna – letting them say yes to more borrowers without taking on more risk on their own books. And the network keeps widening.
This year, the company added a distinctive new partner to its roster of lenders: Upstart, one of the biggest players in the AI lending business. It now taps Pagaya’s engine as well — a sign of how much trust the industry is placing in this particular flavor of AI decision-making.
Pagaya has put together a run of profitable quarters and has been raising its own outlook as that network keeps growing.
It’s a reminder that you don’t need to be the company borrowers or shoppers have heard of to profit from this shift. Sometimes the biggest opportunity is the business working behind the scenes, taking a fee every time the decision gets made.
Lemonade, Oscar Health, and Pagaya prove the same point three different ways. Hand the grunt work to AI, and the business can grow without the payroll growing alongside it.
That’s what makes each of them an AI Cash Machine — the profit shows up because the AI is doing the work, not adding to it.
Next, we turn to companies doing something different with AI — not deciding, but persuading.
II. The Persuaders
For as long as there’s been advertising, someone has had to make a guess.
A cereal company buys a 30-second TV spot because market research suggests that’s when parents are watching. A billboard goes up on a particular stretch of highway because a lot of vacationers pass by. Educated guessing, dressed up as strategy.
The Persuaders in this report don’t guess about a group. They guess about you, specifically — and they make that guess over again, each time you show up, using what you’ve just done rather than what people like you did last year.
Say you pull up a website on your phone. In the fraction of a second before the page even loads, one of these companies has already looked at what you, personally, have clicked on before, what you’ve bought, maybe even the time of day — and picked, out of thousands of ads it could show, the one it reckons is most likely to get you to reach for your wallet. Then it puts that ad in front of you.
A market researcher finds out how well their guess worked three months later, when the next survey comes back — and even then, they’re only ever talking about “people like you,” not you. This machine finds out the second you click, or don’t, and it uses that answer to make its next guess better, for the next person, before their page has even loaded.
This is why so many people are convinced their phone is listening to them. You mention a pair of hiking boots out loud, and an ad for hiking boots shows up an hour later, and it feels like there’s no other explanation.
There usually is one — you’d searched for something similar last week, or a friend with the same interests just did, and the machine noticed. But the guess is good enough, often enough, that “it’s listening” starts to feel like the simpler explanation.
This is a different kind of AI job than the Deciders above. Nobody’s credit or insurance claim depends on the outcome here. But advertisers already spend more than $1 trillion dollars a year trying to get this right, so even a small edge in guessing correctly is worth a fortune to whoever gets there first.
AppLovin (APP) — The AI Media Buyer
AppLovin’s AI engine, called Axon, has one job: decide which ad to show on which phone, in real time, out of millions of possible pairings, across an enormous network of mobile apps and games.
It sizes up a user based on their behavior — what they’ve engaged with, what they’ve bought before, how they use their phone — and picks the single ad most likely to make them stop scrolling and tap “buy.”
To put that in perspective: a human ad-buying team might test a handful of ad variations against a handful of audience segments and call it a good day’s work. Axon runs an almost incomprehensible number of these micro-decisions every single second, continuously learning from which combinations actually convert into a sale and adjusting itself accordingly, all before a person is ever in the loop.
AppLovin built its reputation doing this for mobile game developers, helping them find the players most likely to download and spend money in their games. But it’s recently pushed that same targeting engine well beyond gaming, into e-commerce advertising and connected TV — the ads that show up when you’re streaming a show on your television. Each new category means a bigger pool of advertisers who want access to Axon’s targeting.
The result of all this is one of the most profitable software businesses in the public markets today —because there’s no factory, no inventory, and very little human labor standing between the AI’s decision and the advertiser’s payment.
Worth knowing plainly before you buy: AppLovin is the subject of an SEC investigation into how it collects the data that feeds Axon, following short-seller reports that raised questions about its data-gathering practices.
We think AppLovin still earns its place here as the flagship of the whole “AI runs the business” thesis in this report — the numbers are simply too extraordinary to ignore. But given the live investigation, this is a name to size modestly rather than go all-in on.
Zeta Global (ZETA) — The AI Marketer
Zeta has built one of the largest private databases of consumer behavior anywhere in the world — hundreds of millions of individual people, and trillions of data points describing what they browse, what they buy, and what they seem about to want next.
That database on its own would already be valuable. What makes Zeta interesting as an AI company is what sits on top of it: an AI agent called Athena.
It digests all that behavioral data and tells big-name brands exactly who to target, on which channel — email, text message, an app notification, a display ad — and at what moment in a person’s day they’re most likely to respond.
It’s the difference between a company blasting the same offer to everyone on its list and a company that knows, almost individually, who’s ready to buy right now.
Zeta has also struck partnerships with OpenAI and Palantir. It’s pairing its enormous trove of consumer data with some of the most advanced AI tools available anywhere, to sharpen that targeting even further. For a brand trying to reach the right customer, that combination is hard to replicate from scratch.
Zeta has beaten Wall Street’s expectations and then raised its own guidance again right afterward for 19 straight quarters — a streak that’s hard to find matched anywhere else in this report — all while trading at a cheaper price tag than the rest of the software sector.
That combination — real, proven growth at a price that doesn’t yet assume the story keeps going forever — is what earns Zeta its spot on this list.
Reddit (RDDT) — The AI Forum
Reddit occupies an unusual spot in this report, because it gets paid by both sides of the AI boom at the exact same time — a rare position for any single company to hold.
On one side, advertisers pay Reddit for its AI-powered targeting tools, which decide which product or service should show up in front of which specific online community.
Reddit is organized into thousands of individual communities, each with its own culture, interests, and conversations — sneaker collectors, new parents, home cooks, car enthusiasts. Figuring out which community is the right fit for which advertiser used to require a human media planner guessing from the outside. Reddit’s AI instead reads the conversations happening inside each community and matches advertisers to the audiences most likely to care.
On the other side of the business is something that didn’t meaningfully exist for Reddit a few years ago: selling access to its own enormous archive of real human conversation as training data for other companies’ AI models.
Reddit’s forums contain billions of posts and comments built up over two decades — genuine questions, arguments, recommendations, and opinions from real people. That’s the kind of authentic, human-written material that AI companies like Google and OpenAI need to train their own models, and they now pay Reddit directly for the right to use it.
Put together, Reddit has become a company that profits whether a business wants to reach human customers or a lab wants to teach a model to think more like one.
It’s posted a remarkable run of seven quarters with 60%-plus revenue growth, strong profitability, and unusually high margins for an advertising business — a combination its own CEO has pointed out that almost no other public tech company can currently claim.
III. The Rebuilders
Every industry has its own particular headache.
A restaurant has to guess how much chicken to order before a Friday-night rush. A school has to notice which student is falling behind before it’s too late to help. A physical therapist has to watch someone do an exercise and catch the exact moment their form goes wrong.
The Rebuilders in this report picked one of these ordinary, easy-to-overlook industries and rebuilt it around AI that can do that job at least as well as the person who used to do it by feel — and do it for everyone, all at once, without ever getting tired or distracted.
These are narrower bets than some of the other names in this report. A Rebuilder lives or dies with the one industry it’s chosen to transform. But that narrowness cuts both ways — it also means more room to run if that industry keeps shifting toward AI the way these four companies are betting it will.
Toast (TOST) – The AI Restaurant Manager
Toast is the technology platform running the back office of well over 100,000 restaurants across the country — the point-of-sale terminal at the register, the card reader on the table, the online ordering system, and even the software that runs payroll for the kitchen and waitstaff.
Running a restaurant well used to come down to gut instinct built up over years — a manager who could just tell, from experience, when to order more chicken or when Friday night was about to get slammed.
Toast’s AI layer, called Toast IQ, is replacing that gut instinct with advanced data analytics. It reads live activity across Toast’s entire network of restaurants — every sale, every shift, every ingredient used — and turns all of that into plain, practical instructions for a busy owner who doesn’t have time to study a spreadsheet.
Reorder the chicken before the walk-in cooler runs empty. Add another server before Friday night gets slammed. Nudge a menu price up before a slow night gets even slower. These aren’t hypothetical examples — they’re the kind of everyday, unglamorous decisions Toast IQ makes, at a scale no restaurant manager could track by hand across ingredients, staffing, and pricing all at once.
It’s an AI assistant for an industry that has always run on razor-thin margins, where a single bad week of over-ordering or under-staffing can matter enormously.
Toast is profitable today and has been raising its own guidance inside an industry that most investors would never think to look at for AI exposure at all.
Duolingo (DUOL) – The AI Teacher
Duolingo is a language-learning app with about 137 million active monthly users. And it’s become one of the most interesting AI content factories in existence.
Building a language course used to be slow, expensive, human work — a team of linguists and teachers writing exercises, checking translations, and recording native speakers, one lesson at a time.
Duolingo’s AI now generates entire lessons and course content at a scale no team of human teachers could ever hope to match, writing new exercises, translating them accurately, and even voicing them aloud, automatically and continuously.
That capability didn’t stay confined to languages for long. Duolingo has used the same content-generating engine to expand well beyond its original core, launching courses in math, music, and even chess — each growing into a meaningful business of its own in a remarkably short window of time, something that would have taken years to build the old, human-curated way.
The company is solidly profitable, with a large and growing user base, and its AI-generated content pipeline is exactly what’s letting it move into these new subjects so quickly.
It’s an example of AI not just making an existing business more efficient, but actively helping a company become something bigger than it used to be.
Tempus AI (TEM) — The AI Genetic Detective
Tempus has built one of the largest libraries of clinical and genetic data in the country, and it uses AI to turn that library into something doctors and drug companies can act on.
Here’s the problem Tempus solves. A cancer patient’s genetic makeup can determine which treatments are likely to work and which ones won’t. But sifting through that genetic complexity used to require enormous time and specialized expertise that most oncologists simply don’t have room for in a busy practice.
Tempus’s AI analyzes a patient’s genetic and clinical data and helps their doctor pick the treatment most likely to actually work for that specific patient, rather than a one-size-fits-all standard approach.
It does something similar for the drug companies developing tomorrow’s treatments. Instead of designing a clinical trial around broad assumptions about a disease, drugmakers can use Tempus’s data to identify precisely which patients are likely to respond to a new treatment, making research faster and more targeted than the industry’s traditional trial-and-error approach.
Every test Tempus runs adds one more data point to its library — and every addition makes its overall advantage a little wider and harder for any competitor to copy from scratch, since there’s no shortcut to building a library this large.
Existing hospital and pharmaceutical customers have been spending significantly more with Tempus each year, a strong sign that the product becomes more valuable to a customer the longer they use it.
This is the most speculative stock in the report. Tempus is still investing heavily to build out its data library and its business at large, and it’s the riskiest name on this list by a wide margin.
But if precision medicine is where healthcare is headed — and the direction the industry is moving in suggests that it is — Tempus is one of the purest, most direct ways to bet on that future today.
Hinge Health (HNGE) — The AI Physiotherapy Practice
Hinge Health runs a digital clinic for musculoskeletal pain — the back, joint, and arthritis issues that together make up the single largest category of employer healthcare spending in the entire country, bigger even than diabetes or heart disease.
Traditionally, treating this kind of pain meant in-person visits with a physical therapist, who watches how you move, corrects your form, and adjusts your program week to week.
Hinge has rebuilt that entire experience around a smartphone camera. Its computer-vision technology watches a patient perform an exercise at home and corrects their form in real time, the same way a physical therapist would if they were standing right there in the room.
An AI assistant called Robin handles much of the coordination work that used to require a human care team — checking in with a patient, adjusting a program as they improve, answering routine questions — freeing up Hinge’s actual clinical staff to focus on the cases that genuinely need a person’s judgment.
Employers and health insurers pay Hinge a fee per covered employee, and the company already counts a majority of the very largest employers in the country among its clients.
Hinge is profitable, growing sharply, and has raised its full-year outlook twice in recent months — with leadership pointing specifically to AI-driven efficiency, rather than simply hiring more staff, as part of the reason margins keep improving. It’s the cleanest example in this entire report of an old, in-person industry rebuilt from the ground up around AI.
A Word on Risk
Now that we’ve walked through all 10 AI Cash Machines, it’s worth taking a moment to talk about risk.
These companies span a wide range of sizes, and that matters. To make the risk profile of each pick clearer, here’s how they break down by size and volatility.
Large-Cap Companies (Lower Volatility, Established Businesses)
- AppLovin (APP)
- Reddit (RDDT)
- Toast (TOST)
These are large, well-established businesses with real, proven revenue. None of them need the “AI Cash Machine” thesis to already be working — it already is, and each one has years of profitable operating history behind it. For most investors, especially those newer to thematic investing, these are the lowest-risk way to get exposure to this theme.
Mid-Cap Companies (More Direct Exposure, Bigger Swings)
- Lemonade (LMND)
- Oscar Health (OSCR)
- Zeta Global (ZETA)
- Duolingo (DUOL)
- Tempus AI (TEM)
- Hinge Health (HNGE)
This is the biggest group in the report, and for good reason — it’s where the AI Cash Machine story is often still being written in real time. These companies are more exposed to the fortunes of one particular industry, or to a handful of big customers, or to a single regulatory decision.
That gives them more room to run if things break their way, but it also means their stocks can move sharply on a single quarter’s results or a single piece of news.
One name in this group carries meaningfully more risk than the rest: Tempus AI, which we flagged earlier as the report’s most speculative pick. It’s still spending heavily to build its business and remains further from turning a profit than anything else on this list. Treat it more like the small-cap below than like its mid-cap neighbors when you’re sizing a position.
Small-Cap Companies (Speculative, High Volatility)
- Pagaya (PGY)
This is the smallest, most speculative name on the list. It has the least trading history, the thinnest cushion if something goes wrong, and the most potential to multiply in value if its bet on AI-driven lending keeps paying off. Size a position here accordingly.
Practical Risk Management
A few simple principles can help manage risk across a list like this one:
- Start with the large caps. They offer real exposure to the AI Cash Machine theme without requiring every part of the story to play out perfectly.
- Size small with the speculative names. A smaller company should typically make up a smaller position in your portfolio — not your core holding.
- Expect volatility, and don’t let it shake you out. Sharp price swings are normal for growth companies riding a young trend. Volatility is the price of admission for outsized gains.
- You don’t need to own all 10. This report is a menu, not a checklist. Even one or two well-chosen picks can give you real exposure to this theme.
The AI Cash Machine trend is still early. Matching your position sizes to your own tolerance for risk is just as important as picking the right companies in the first place.
From AI Cash Machines to AI-Powered Returns
The AI revolution isn’t just changing how the world works. It’s also changing how we invest.
At TradeSmith, we’ve built our own AI system — not for designing chips or cooling data centers, but for predicting short-term stock moves.
We call it Predictive Alpha.
It’s a large-scale AI model trained on vast amounts of stock market data. And it uses proprietary machine learning models to forecast the expected price path of thousands of stocks for every trading day over the next month.
We’ve engineered Predictive Alpha on more than 120 million data points. These include…
- 4.2 million historical price outcomes, spanning seven years and more than 2,300+ stocks
- 88.9 million daily forecasts, covering 21 forecast days for every stock on every trading day of the year
- Plus, tens of millions of additional “validation runs,” including target accuracy, pattern recognition layers, and more
Based on this data, Predictive Alpha learns from the past, adapts to the present, and projects the future.
We’ve found that every stock has a specific “profit window” when it can move the most. For example, Tesla may have a 6-day window, but Apple may have a 15-day window.
Every day, our system finds the ideal window to trade a particular stock.
Predictive Alpha can’t predict the future with 100% accuracy. And it won’t get every trade right. But it can forecast, with up to 85% probability, where stocks will be tomorrow.
And the longer it runs, the smarter it gets. Here are the top 10 wins that it has identified so far…
- 9.5% on SOFI in 3 days
- 9.6% on SOFI in 8 days
- 10.3% on UPST in 1 day
- 11.4% on MCW in 3 days
- 11.6% on ANF in 3 days
- 12.0% on ACCO in 26 days
- 12.6% on SLV in 4 days
- 16.4% on APLD in 6 days
- 17.6% on NTAP in 21 days
- 25.5% on CVNA in 2 days
These aren’t annual returns. They’re happening over a matter of days. Repeating these types of gains over these timeframes is like having a “house edge” on a casino-hopping Las Vegas trip.
Put the “House Edge” on Your Side
Each year, the Nevada Gaming Control Board (NGCB) writes a report on the success of the casinos on the Las Vegas Strip.
We all know the house always wins — this report tells us by how much.
In 2019, tens of millions of people flocked to Las Vegas to play casino games like blackjack, poker, and roulette. The NGCB shows that, collectively, those folks went home $6.6 billion poorer.
The house edge is the percentage of a player’s bet that the casino is likely to win. Said another way, it’s the statistical advantage a casino holds in any given game.
Casinos don’t win year in and year out because they get lucky. They have a deep knowledge of probabilities. And they use this knowledge to build an edge into every game they operate.
They make small wins thousands of times a day, millions of times a year. And these small edges, applied relentlessly, pile up.
It’s the same with Predictive Alpha. A 9.5% win in three days is impressive. Sustained across a trading year, those kinds of wins really add up.
The secret to making money in Vegas or in markets is simple: find your edge and apply it over and over. Predictive Alpha gives you that edge.
Want to See Predictive Alpha in Action?
With Predictive Alpha, we didn’t chase the impossible dream of predicting the future or being right 100% of the time. We looked for an edge we could exploit over and over again.
Billionaire casino operators know how powerful that kind of edge is. The world’s best traders know it as well.
And as I mentioned up top, you can now put this edge to work for you.
So, if you haven’t already, check out a live demo.
You’ll see how our AI analyzes five top stocks and predicts where their share prices could land over the next 21 days. It’s one of the most powerful trading tools ever developed — and you can try it free.
Just click the link below. It will take you to the page where you can try out the free demo, live on screen.
When you do, you’ll also discover how you can use this same AI system to forecast price movements on any of more than 2,300 stocks we track. From the AI infrastructure winners in this report to every major stock on Wall Street.
The future belongs to those who embrace AI — as an investment theme and as a trading edge.
Keith Kaplan
CEO, TradeSmith