The $1.8 Billion Company With Two Employees

Everyone's been yelling about this all week.

You've seen it. The New York Times runs a profile on some guy named Matthew Gallagher who built a $1.8 billion telehealth company with his brother and a handful of AI tools.

I read the article. Then I read the Futurism investigation. Then I asked Claude to give me a review of the class-action lawsuit. Then I spent one day going through with Claude Cowork fetching every source I could find, customer reviews, FDA warning letters, affiliate marketing programs, the founder's prior company, the partner companies behind the scenes. All of it (I have burst some thons of token here... 🔥).

And here's what I think: everybody is arguing. Nobody is learning.

So let me break it down for you. Because there is a real blueprint buried under all the noise. One that changes how you should think about building businesses with AI. And it has nothing to do with telehealth or weight loss drugs.

Matthew Gallagher. 41 years old. Los Angeles.

And no, he's not some Silicon Valley wunderkind who stumbled into a billion dollars. His story is actually wild:

What did work was e-commerce. Before Medvi, he built Watch Gang, a watch subscription box. It took off fast:

Here's the part that matters: Watch Gang never turned a profit.

Gallagher said later that having 60 employees:

That lesson is the entire origin story of Medvi. Not the AI tools. Not the GLP-1 wave. The painful realization that more people doesn't equal more progress. He'd already learned the expensive way what most founders never learn.

Keep that in your head. It's going to matter.

I'm going to be direct because this is where most people get confused.

Medvi is not a healthcare company. Let me be clear about what it doesn't do:

Medvi is a brand. And its specific market is in the likes of Nutra Affiliate Marketing or the Nutra Vertical (what is very, veeeery popular in brazil, or by brazilians 🇧🇷, for the good or not).

Mdvi = A really good brand + Sitting on top of other people's infrastructure.

Here's how it actually works. Two companies do the heavy lifting:

These companies spent years and millions building plug-and-play telehealth systems. They welcome partners like Medvi because every customer Gallagher brings is revenue for them too.

Gallagher's piece? The website. The checkout flow. The ad creatives. The customer service chatbot. The analytics dashboard. All built with AI tools. He owns the customer relationship and the marketing engine. That's it.

Here's your homework before you keep reading: open a new tab and search "[your industry] + white label provider" or "[your industry] + API." Spend 30 seconds. Just see what comes up. I'll wait.

The customer experience is simple: you go to Medvi's site, fill out a health questionnaire, a licensed doctor from OpenLoop's network reviews it, and if you qualify, they prescribe a compounded GLP-1 weight loss drug. A partner pharmacy compounds and ships it. Monthly subscription. Medvi takes its cut.

Now. If this model sounds familiar, it should.

This is the exact same model as:

The difference? Gallagher did it with two people and $20,000 in AI tools.

I'm going to say this differently because it's the most important sentence in this entire piece:

The infrastructure already existed. The demand already existed. AI just eliminated the cost of building the bridge between them.

Read that again. Seriously.

Here's the timeline. Try not to fall off your chair:

His only employee? His younger brother Elliot, 36, based in Cincinnati. Hired to handle communications so Matthew could focus on growth.

Sam Altman emailed the New York Times saying he'd won his bet with fellow tech CEOs about when a one-person billion-dollar company would appear. Said he was very eager to meet the person who pulled it off.

When I first saw these numbers I thought they were fake. They're not. The New York Times verified the financials.

And when you understand the GLP-1 market, it actually makes sense:

The pricing gap is enormous. The demand is insane.

Gallagher didn't create demand. He surfed a tsunami that was already coming. He just built the fastest board.

Here's every tool and what it did. This is the part I found most interesting as a builder. I went deep on this one with Claude helping me map it all out.

Total setup cost: about $20,000.

Here's what blew my mind. At $401 million in revenue, he still didn't hire anyone.

The AI kept running. Claude handled support. Ad creatives kept generating. Analytics kept monitoring.

The chatbot hallucinated wrong pricing at one point. He just honored the incorrect prices and kept going. Because fixing a chatbot bug was still cheaper than hiring a customer service department 😅

That's the part that should make every builder reading this sit up in their chair. Not the $1.8 billion. The fact that AI held together at $401 million in revenue with two humans overseeing it.

This is the distribution layer that made the whole thing scale. And nobody in the AI discourse is breaking it down.

The math is beautiful:

This is the same playbook that every successful e-commerce brand uses. He just ran it with AI creating the ads, AI handling the customer service, and AI tracking the analytics. The affiliate network did the distribution.

Gallagher himself said it:

He's right. And that quote tells you everything about what the real moat is. It's not the technology. It's not the AI tools. It's who gets to the customer first with the best funnel. That has always been true. AI just makes it truer.

I'm going to give you the full picture. That's what we do here. No gloss. No cheerleading. No hating either. Just facts.

Fake before-and-after photos. Futurism investigated Medvi in May 2025 and found the company used AI deepfake technology to create fake patient transformation photos. They traced a photo of "Michael P" to a real Reddit user's weight loss journey from 2016. The guy lost 35 pounds in one year. Medvi:

When caught, they swapped the photo for a different person but kept the same fake stats. These photos were still there as recently as March 2026.

Fake AI doctors running ads. Meta's ad library shows AI-generated physicians promoting Medvi: "Dr. Sarah Martin," "Dr. Monica Ashford," "Dr. Lena Fischer." None of them are real people. Just AI slop in video form selling weight loss drugs. Real doctors whose names appeared on Medvi's website told Futurism they had nothing to do with the company. Medvi eventually added a disclaimer saying people in their ads "may be actors or AI portraying doctors." Yeah.

FDA warning letter. February 20, 2026. Warning Letter #721455. Violations:

This wasn't Medvi-specific. But they were named.

Class-action lawsuit. Filed November 2025 against OpenLoop Health and compounding pharmacy Triad Rx, the infrastructure partners behind Medvi. The core allegation: compounded oral tirzepatide tablets are "modern-day snake oil."

The science behind it:

That's not a marketing problem. That's a "does the product even work?" problem.

Customer complaints. On Consumer Affairs: 704 reviews. 51% are one star. Common themes:

On Trustpilot: 4.4 stars with 11,498 reviews. Make of that difference what you will.

I'm not sharing this to pile on. I'm sharing it because you can't learn from a case study you only see one side of.

The business model is real. Full stop.

Hims & Hers does the exact same thing at $2.35 billion. Nobody calls them a scam. The telehealth-to-compounded-drugs model is a legal, established, multi-billion dollar category.

The marketing practices are where things get seriously wrong. Fake testimonials. Deepfaked photos. AI-generated fake doctors. The New York Times framed this as "faking it till you make it."

I don't buy that framing. Fake doctors selling health products isn't hustle culture. It's deception.

The oral tirzepatide question is the most serious one. If the science is right and these tablets don't actually absorb into the bloodstream, then real people paid $279.99 per month for pills that don't work. That's a different conversation than aggressive marketing.

Here's my honest take:

Because here's the thing. Hims & Hers got to $2.35 billion without deepfaked doctors. The market is that big. The demand is that real. The model works when you do it right. The shortcuts weren't necessary. They were just faster.

My vereditc?!! Fake Naaaaaty (In this case: fake first 1 Billion Person busines)

Forget the ethics debate for a second. Let me tell you what I see as a builder.

1. AI didn't create a new business. It eliminated the cost of entering an existing one.

Medvi's innovation wasn't the product or the medical model or even the technology. It was the realization that the only thing preventing a solo operator from competing with 2,400-employee companies was the cost of building the customer-facing layer. AI zeroed that cost out.

This applies to every industry where infrastructure exists but sits behind a wall of expensive front-end operations:

Wherever there are white-label back-ends waiting for distribution, AI just removed the front-end cost.

2. The "wrapper business" is the new dropshipping. But better.

Gallagher built a brand on top of CareValidate and OpenLoop, the same way dropshippers build brands on AliExpress supply chains. But this version is more powerful because the infrastructure providers are institutional-grade companies with real compliance, real physicians, and real regulatory frameworks.

Think about what the equivalent is in your world. Who are the infrastructure companies in your target industry? The ones providing back-end services that end consumers never see? Your job is to build the front end that connects them to customers better than anyone else.

3. Distribution is still the only moat.

Every builder has access to the same LLMs, the same image generators, the same code assistants. The only differentiator is your ability to reach people. Gallagher is an e-commerce marketer who found the highest-demand product in America and built the fastest funnel to sell it. AI was the vehicle. Distribution was the engine.

4. The biggest opportunities might not be software.

This is for you that thinks AI will be an enabler only in the digital world. I still believe that's the best path for most of you. But Medvi is a reminder that AI doesn't care about industry labels. The same tools that let you build a subscription app can let you build a telehealth brand, an insurance brokerage, a real estate matching service, or a financial planning platform.

The question isn't "what software can I build?" It's "where is there massive demand, existing infrastructure, and a customer-facing gap that AI can help me fill?"

5. Speed is the whole game.

Gallagher went from idea to $300K in revenue in one month. Not because he's a genius. Because AI compressed the build cycle from months to days and the market was waiting. Stop optimizing your MVP. Stop debating tech stacks. Stop polishing your landing page. Ship it. The market will tell you what to fix.

Step 1: Find a market with explosive demand and proven willingness to pay. Gallagher didn't create demand. He rode a wave. Your job is to find where demand already exists and is underserved.

Step 2: Find existing infrastructure you can plug into. CareValidate and OpenLoop built the hard stuff. Gallagher just connected. Look for the white-label equivalent in your target industry.

Step 3: Use AI to build the entire customer-facing layer. Website, copy, ads, support, analytics. What used to cost $500K and require 30 people now costs $20K and requires one person with Cursor, Replit, or Bolt.

Step 4: Invest everything in distribution. Gallagher spent his energy on affiliate marketing and ad creatives. The product was someone else's. His job was getting it to people faster than anyone else.

Step 5: Stay lean. Then stay leaner. Two employees at $1.8 billion in revenue. The Watch Gang experience taught him that every hire is a cost and a bottleneck. Only add humans when AI literally cannot do the job.

If Gallagher had done everything the same way but without the fake testimonials, without the deepfaked photos, without the AI-generated doctors: would it have worked?

Probably yes. Not at the same speed. But the market is that big. The demand is that real.

The shortcuts weren't necessary. They were just faster. And that's the part you should refuse to copy.

Build fast. Ship fast. Market hard. But keep it real. The customers are real people spending real money on products that affect their health. That matters more than your growth rate.

Matthew Gallagher proved something that should change how every solo builder thinks about opportunity.

He also proved that when you move that fast with that little oversight, things go wrong. FDA warnings. Lawsuits. Fake doctors. Customers charged for products that might not work.

Both things are true. The opportunity is real. The risks are real.

The next person who builds a billion-dollar company with two employees and AI is going to do it the right way.

It could be you. And you'll be ready because you're here. Hit reply and tell me:

I read every reply.

See you next week. Go ship.

— Bissuh

The DTC beauty category is crowded. To break through, Jennifer Aniston’s brand LolaVie, worked with Roku Ads Manager to easily set up, test, and optimize CTV ad creatives. The campaign helped drive a big lift in sales and customer growth, helping LolaVie break through in the crowded beauty category.

Learn more

Sources and further reading: