This $120M AI Founder Says You Can Make $1 Million in One Week Using These 3 AI Opportunities

Three powerful AI opportunities for making serious income right now include AI video content creation for businesses, building and marketing AI-powered apps, and selling services or infrastructure that directly supports AI companies — and according to Roy Lee, the $120M CEO of Cluey, any one of these lanes can realistically generate a million dollars within months, with the lowest-effort entry point being AI video ads made for companies using tools already available today.

👉 Start a 1-Person Business With Claude AI — Free Quick-Start Guide

Who Is Roy Lee and Why His Words on AI Opportunities Matter Right Now

Picture a 20-something founder sitting in a plain t-shirt, not shaving, talking casually — and somehow running a company valued at over $120 million.

That is Roy Lee, the CEO of Cluey, a software company that now operates seven different products under one roof, all maintained by just two full-time engineers.

Roy built his name by publicly baiting Columbia University into kicking him out, using the attention to launch his company, and turning that moment into millions.

He has since appeared in rooms with partners at Andreessen Horowitz, attended LP summits in Las Vegas, and connected with some of the youngest AI billionaires being minted today.

He is not a distant theorist talking about AI from a conference stage.

He is actively building, actively testing, and actively watching where the AI opportunities that no one is loudly discussing are pulling in real money.

When Roy speaks about what is working in AI right now, the words carry the weight of someone who has already collected the receipts.

And in 2026, what he is pointing at has the potential to change the financial trajectory of any person who moves on it fast enough.

The World Has Changed and Most Businesses Have Not Caught Up

Before diving into the three specific AI opportunities Roy Lee has identified, it helps to understand the core insight that everything else rests on.

Roy describes it plainly: the way to make money with AI today is to capture value from people and companies who do not yet know how powerful AI has become.

He estimates that 99.9% of companies in the world today are sitting on AI use cases they are not acting on.

They are not making AI-generated ads.

They are not using AI to build websites or product demos.

They are not automating things that their competitors are starting to quietly automate.

This gap between what AI can do and what most businesses are actually doing is where the money is sitting right now — wide open, unclaimed, and growing with every week that passes.

If you can position yourself to bridge that gap, even in one narrow way, the business case is already made.

You are not trying to convince someone that AI matters.

You are simply showing up with a result and asking them to pay you to keep producing it.

This is the setup that makes all three AI opportunities Roy points to so immediately actionable for anyone starting from zero.

👉 Get Access to: The AI Blog Monetization Quickstart Guide

The 3 AI Opportunities Roy Lee Says Are Wide Open in 2026

AI Opportunity #1 — AI Video Ads for Businesses

The first and most accessible AI opportunity that Roy Lee identifies is AI video content creation, specifically in the form of video ads made for companies using AI video generation tools.

He describes this as the lowest-effort, highest-leverage entry point available to anyone who wants to start making real money with AI today.

Here is the specific model Roy outlines: you go find a hundred software companies that have raised over ten million dollars in venture capital funding, which means they have a real budget and a real pressure to perform on marketing.

You use an AI video generation tool — Roy specifically names Kling 2.5, which is a real, widely available AI video platform — to create a compelling, scroll-stopping ad for each company, completely free of charge.

You then email each company every single day with the ad attached and a simple message: I made this for you for free, run it as an ad, if it is profitable put me on a retainer for ten thousand dollars a month, if it is not profitable then ignore this email completely.

You repeat this process until you have eight to sixteen companies paying you ten thousand dollars a month each, and you have crossed the million-dollar threshold in annual revenue.

Roy says this is not theoretical.

He says this is something he himself would sign off on as a genuine business model, and he points out that marketing agencies already exist today that subcontract creators — including teenagers who have grown up on TikTok and Instagram Reels — to produce AI video ads at scale, then sell those ads to companies for a premium.

The skill that makes this work is not technical.

It is the simple ability to watch a video and know whether it is boring or not, which is an instinct that anyone raised on short-form content has already developed after years of scrolling.

Roy’s point is blunt: a forty-year-old Facebook user will share the most generic, lifeless AI-generated content without blinking, because they genuinely cannot tell the difference — but a person who has spent years consuming TikTok and Instagram Reels has trained their eye to know what stops a scroll and what gets swiped past.

That eye is the only real qualification needed to start this business.

A useful way to research what already works: look up any company, go to the Meta Ad Library, find their longest-running ad — the one that has been running the longest is almost always their best performer — and remake it with Kling 2.5 as your starting creative template.

Then bring that ad to their direct competitors, because those competitors need the same result and have no existing contract with you.

This is one of the cleanest AI opportunities in the market right now, because the barrier to entry is creative instinct and a free email account.

👉 Get Access to: The AI Traffic Vault

AI Opportunity #2 — Building and Marketing AI Apps

The second of Roy’s identified AI opportunities sits one level above the video ad model, and it requires slightly more ambition but unlocks a ceiling that is far higher.

Roy Lee’s company, Cluey, currently runs seven distinct software products under one corporate umbrella, all maintained by two full-time engineers — and he notes that nearly every single one of those seven products generates over a million dollars a year on its own.

What makes this possible in 2026 is that building a functional, polished software product no longer requires a team of twenty engineers working for six months.

Roy says a single engineer today can build in an afternoon what previously would have taken months and a large team, because modern AI coding tools have closed most of the gap between a novice developer and a senior one.

This creates an obvious AI opportunity for any entrepreneur: identify a specific problem that a specific group of people has, build an AI-powered app that solves it, and market that app using AI-generated content including AI influencers and AI video.

Roy describes the theoretical arc: you build an app in an afternoon, submit it to the App Store, wait the standard review period of a few days, and simultaneously build a marketing system around AI-generated content that runs on autopilot.

He says it is genuinely possible to generate a million dollars in revenue within a week if the product and the marketing land correctly — not as a guarantee, but as a realistic ceiling that the current environment supports.

The person who wins in this space is not necessarily the best engineer.

It is the person who identifies an underserved problem fast, builds a minimum viable solution quickly using AI coding assistants like Claude, and gets in front of the right audience before someone else fills the gap.

Roy also makes a point about competitive density: most companies are so focused on their primary product that they are unwilling to spin up seven parallel products and see what sticks.

That focus creates an opening for the builder who is willing to move fast across multiple ideas.

If you have been thinking about building something but kept waiting until you had more technical skill, Roy’s point is that the skill gap has effectively already been closed by AI — the only thing left is to start.

For anyone ready to take that first step, 👉 Start a 1-Person Business With Claude AI — Free Quick-Start Guide is a free resource that breaks down how to begin building a one-person digital business using Claude AI as your core operating system.

AI Opportunity #3 — Building for the AI Labs Themselves

The third AI opportunity Roy Lee names is the highest-ceiling one, and it is aimed at the most technically capable or well-connected people in the room.

Roy describes the flow of capital into AI infrastructure right now as genuinely staggering — billions being deployed by venture capital firms into companies that are building the scaffolding, data pipelines, hardware optimization layers, and compute infrastructure that the major AI research labs depend on.

The AI labs — companies like Anthropic, OpenAI, Google DeepMind, and others — are not building everything themselves.

They rely on a growing ecosystem of vendors, contractors, and specialized companies to supply them with the things they need to run their models: training data, specialized chips, GPU optimization tools, evaluation frameworks, safety testing infrastructure, and more.

Roy’s argument is that if you have a specific technical skill or domain expertise, the fastest path to building a genuinely large company may not be a consumer app — it may be building something that one of the AI labs would pay for directly or acquire entirely.

He describes cases where founders have woken up to a four-hundred-million-dollar contract from a research lab essentially overnight, not because they had spent seven years grinding, but because they positioned themselves in the right place at the right time with a specific capability the lab needed.

This is the most capital-intensive and technically demanding of the three AI opportunities Roy identifies, but it is also the one with the shortest timeline from start to outcome if the fit is right.

The venture capital environment in 2026 is uniquely open to backing early-stage companies in this space, and Roy’s read is that the window for getting in at the infrastructure level is still genuinely early even now.

For a creator or digital entrepreneur who is further from this space, the actionable takeaway is simpler: pay attention to what the AI labs are publicly saying they need, what papers they are publishing about their bottlenecks, and what categories of tools they are funding or acquiring — because that map tells you exactly where the money is being directed.

Why Roy Lee Says Getting Rich in 2026 Is Painfully Simple

Roy does not soften this point.

He says that compared to any other period in human history — including the early internet, the industrial revolution, or the post-war economic boom — the current environment offers more accessible paths to wealth than have ever existed before, and the only requirement is that someone actually tries one of them.

His argument is not that success is guaranteed.

It is that the ratio of viable opportunity to effort required has never been more favorable for someone starting with nothing.

He points out that a person with no money, no connections, and no technical background can open a laptop today, prompt an AI model to tell them how to make ten thousand dollars this month, get a genuinely useful and specific answer, set up an automation to handle most of the process, and still have only spent a few hours of their own time.

The gap between the version of a person who has made money with AI and the version who has not, Roy says, is simply that the first version actually tried.

This is not motivational poster language.

It is a structural observation about what AI tools have done to the execution cost of starting a business.

Execution — the hardest and most expensive part of building anything — has been compressed by AI to a fraction of what it cost in time, money, and skill just five years ago.

The people winning right now are not necessarily smarter.

They are simply moving while everyone else is still deciding whether to move.

If you are mapping out your first move in this space, 👉 Get Access to: The Medium Mastery is a compact, focused guide on building consistent blog traffic through Medium — one of the most reliable free distribution channels for any AI content business in 2026.

The AI Video Gold Rush Is Already Happening — Most People Just Can’t See It

One of the clearest signals that the AI video opportunity is real and not theoretical is that the infrastructure for it already exists and is being used by real businesses at scale right now.

Roy Lee’s own company, Cluey, currently works with over one thousand creators on a performance-based model: creators make video ads for Cluey’s products, and if those ads perform as paid advertisements, the creators get paid.

He says the company has the capacity to scale that number to ten thousand or even one hundred thousand creators with no structural ceiling, because the math only works in everyone’s favor as long as the ads keep performing.

This is the same model that Roy is describing as an opportunity for individuals — except instead of being one of the thousand creators working for Cluey, you could be the person who builds the agency that connects creators to companies and takes a margin in the middle.

The social media platforms where AI video content is winning right now are Instagram Reels and TikTok, which together represent where cultural moments are created, where brands are discovered, and where individual creators have built eight-figure businesses from a phone and a clear point of view.

AI video tools like Kling 2.5 now allow someone to generate realistic, scroll-worthy UGC-style video content without appearing on camera, without a production budget, and without prior video experience.

The creative ceiling for these tools is rising every few months.

The person who learns to use them well today will be three skill-levels ahead of the person who waits until they are fully mainstream.

Roy’s take on competitive advantage in this space is also worth noting directly: the average forty-year-old marketing decision-maker at a mid-sized company does not have a trained eye for what makes video content stop a scroll.

They will approve content that a TikTok-raised creator would immediately identify as flat and boring.

That gap in taste — between the buyer and the creator — is where the freelancer with good creative instincts can position themselves as genuinely irreplaceable.

For anyone building a content business around these AI opportunities, 👉 Get Access to: The Flipboard Traffic Workflow Kit covers how to use Flipboard as a traffic amplification layer that consistently drives readers to your content from outside the usual social media channels.

How Roy Lee’s Company Operates Differently From Every Other Startup

It would be a mistake to look at Roy Lee’s success and chalk it up to luck or timing without examining the specific structural decisions he made that most companies are still not making.

The first is the low-engineering headcount.

Cluey runs seven products on two full-time engineers.

Roy’s position is that AI coding tools have closed the skill gap between a junior and senior engineer to the point where headcount no longer correlates to output the way it once did — and the companies that have not updated their hiring models around this reality are already losing ground to those that have.

The second is multi-product thinking.

Most startups are single-product companies by culture and philosophy.

Roy’s company treated additional products not as distractions but as revenue experiments — each one is a separate bet, most of them profitable, and the ones that are not are small enough to run at minimal cost while the ones that work generate seven figures annually.

This thinking applies directly to the solo creator or solo entrepreneur reading this article: you do not have to commit to one digital product, one audience, or one income stream.

Building multiple products using AI tools that lower the cost of production means that diversification is now accessible to a one-person business in a way it never was before.

The third differentiator Roy names is creative volume.

While most companies are still trying to find the one perfect ad or the one perfect content piece, Cluey is running a volume-based creative model — thousands of videos from thousands of creators, with the algorithm deciding which ones win.

This is a mindset shift as much as it is a strategy.

The creator who makes fifty AI video ads and sends them to fifty different companies will find a client faster than the creator who spends three weeks perfecting one ad before sending it to anyone.

Volume is the unfair advantage when AI has made production cheap.

👉 Get Access to: Start a 1-Person Business With Claude AI walks through the full framework for building a lean, AI-powered solo business with multiple income streams — built specifically for the creator or entrepreneur who wants to move fast without a team.

What Happens to People Who Wait

Roy Lee is not gentle about what he sees on the other end of hesitation.

He observes that the average competitor in the market right now is lazier than at any previous point in history.

Not less intelligent — just less willing to act.

He describes people who have access to AI tools that would answer the specific question of how to make money this month, who could automate most of the execution, who could run the entire process from a free account on their phone — and who still are not doing it.

His framing is blunt: if you are not rich right now, or not genuinely close to it, or not running something with a real chance of getting there, then the evidence suggests you simply do not want it badly enough to act.

That is not a judgment — it is an observation about how low the cost of entry has become.

When starting a business required capital, connections, geography, and years of experience, the excuse of not having those things was legitimate.

That excuse has been substantially eliminated by AI tools.

The three AI opportunities Roy identifies do not require funding, a degree, a location, or a prior track record.

They require creative instinct, a willingness to send cold emails or generate content, and the consistency to keep showing up until something clicks.

The person who reads this and takes one action today — even a small one, even just drafting a pitch email to one company — is already ahead of the majority of people who will read the same information and close the tab.

👉 Get Access to: The AI Traffic Vault

Your First Step Into These AI Opportunities

The clearest starting point, based on everything Roy Lee describes, is AI video ads — specifically because it requires the least upfront infrastructure, can be started today, and has a defined path from first video to first paying client.

Here is that path laid out simply:

Pick one AI video tool that is currently accessible — Kling 2.5 is the one Roy names directly, and it is a real platform with a free tier available for testing.

Find ten to twenty software companies that have raised significant venture capital funding — this information is publicly listed on Crunchbase, which is a free database of startup funding rounds.

Go to the Meta Ad Library, search each company’s name, and find their current or recently running ads to understand what creative style they are already responding to.

Use Kling 2.5 to create one AI video ad for each company — something you would actually stop and watch yourself if it appeared in your feed.

Email each company with the ad, no invoice attached, with a simple offer: run this, if it performs, pay me for more.

Follow up every day until you get a reply.

Eight clients at ten thousand dollars a month each equals nearly a million dollars in annual revenue.

Roy says it himself — this is a model he would personally sign off on.

The window for being an early mover on these AI opportunities is not permanently open.

Every month that passes, more people discover these models, more competition enters, and the gap between the informed mover and the late arrival narrows.

The best time to start is now.

👉 Get Access to: Start a 1-Person Business With Claude AI

We strongly recommend that you check out our guide on how to take advantage of AI in today’s passive income economy.