The AI Bubble Explained: Why Your Writing Assistant Might Disappear

Part 2 of 3: Understanding the rot economy before it’s too late

Last week, I told you why mature writers need to pay attention to the AI bubble. Today, I’m going to explain what this bubble actually is—the companies promising to revolutionize your writing process might not be around in five years. Or five months.

Pour yourself something strong. This gets weird.

The restaurant that cares more about expansion than food

Imagine a restaurant chain that keeps opening new locations even though every single one loses money. The food? Getting worse. The prices? Going up. The portions? Shrinking.

But investors love it. Why? Because “number go up” is all that matters. More locations, more revenue, more market share. The fact that the business model is fundamentally broken? Meh.

Tech critic Ed Zitron calls that the “rot economy.” Companies prioritize growth above everything else—product quality, customer satisfaction, even making actual money. The metrics investors care about (users, revenue, market share) must constantly increase, no matter the cost.

In a rot economy, appearing innovative to investors becomes more valuable than actually creating useful products for real people.

Sound familiar? It should. Because this is exactly what’s happening with AI companies right now. You’re seeing it.

The math that doesn’t math

Let me give you some numbers that should make you sit up and notice.

OpenAI—the company behind ChatGPT—burned through $9 billion so they could lose $5 billion in 2024. Read that again slowly. They spent $14 billion and only took in $5 billion from customers. That’s like spending $140 to make $50.

No business can survive that way. Period.

But it gets weirder. In summer 2025, OpenAI announced completely different revenue numbers two days apart—$12 billion “annualized” on July 30, then $13 billion “annualized” on August 1. You can’t have two wildly different predictions in the same week unless you’re playing games with how you calculate things.

Other AI companies? Same story. Perplexity spent 164% of its revenue on computing costs in 2024. They spent $164 for every $100 they made from customers. That’s before paying employees, rent, or anything else.

Does that sound like sound business development?

The acquisition that never happened

Here’s something that should tell you everything: we’re three years into the AI bot-com boom, and almost no real acquisitions have happened.

Think about that. When a startup creates genuine value, bigger companies buy them. Instagram sold to Facebook—$1 billion. YouTube sold to Google —$1.65 billion. WhatsApp sold to Facebook—$19 billion.

But in AI? Crickets. The only “acquisitions” have been weird licensing deals where Microsoft or Google pays hundreds of millions to hire executives and license technology—but doesn’t actually buy the company.

The only genuine AI company acquisition anyone can find? NICE buying Cognigy for $955 million. That’s one real acquisition in three years.

If AI is truly transformative, why won’t Google, Microsoft, Amazon, or Meta buy these companies at their supposed valuations? Cursor is valued at $9.9 billion—why hasn’t anyone bought them?

The answer seems to be because those valuations aren’t real, and I’ll explain that to you in my published report.

The trap nobody can escape

Here’s where it gets really dark. AI companies can’t fix their cost problems without destroying themselves. Isn’t that what happened to Hal and Dave?

Say you’re running an AI company valued at billions because you promised investors explosive growth. To actually make money, you’d need to raise prices and slow expansion. But the moment you announce that plan, your valuation collapses because you’re no longer promising growth. Now you can’t raise money to stay in business.

It’s worse than that. Even if one company decided to fix its costs, competition prevents it. No individual company can slow down without being eliminated by competitors.

So what’s the bubble?

The bubble is this: AI technology has genuine value and will be around forever. But the companies providing it right now—the startups you’re paying $20 or $200 per month—are built on unsustainable economics.

Real value and an over-inflated bubble can coexist. Amazon and Google emerged from the dot-com crash as titans. But Pets.com and Webvan—which raised and lost hundreds of millions—vanished overnight.

The bot-coms will follow the same script. The technology stays. Many of the companies providing it will fail spectacularly. A select few will survive and become the AI titans of the next bot generation.

I think I can probably predict who those will be. You need to get a copy of The AI Rot Economy, so you know, too.

What this means for you

If you’re building your writing business on top of a startup AI platform, like Claude, ChatGPT, or Perplexity, you’re building on quicksand. When that platform goes under—and many will—what happens to your workflows? Your prompts? Your systems?

The companies rushing to sell you AI writing tools, AI content generators, AI marketing assistants? Most of them are burning investor cash at an unsustainable rate. They’re in a race to grab market share before the money runs out. And it will run out.

You need to understand this landscape. Not to panic. Not to avoid AI entirely. But to position yourself strategically so when the bubble bursts—and it will—you’re still standing.

Because here’s the thing mature writers understand that younger entrepreneurs don’t: we’ve seen this before. We know how it ends. And we know how to build something that survives the crash.

The strategy you need (but won’t find here)

The full report—The AI Rot Economy: Why Your AI Assistant Is a Ticking Time Bomb—lays out exactly what you need to do. Which platforms to focus on. Which ones to avoid. How to build AI-augmented workflows that survive the consolidation. How to become AI-proficient without becoming AI-dependent.

I explain the valuation trap in detail. The competitive prisoner’s dilemma that prevents any company from fixing its problems. The infrastructure lock-in that’s already costing companies billions. The talent cost structure that makes profitability mathematically impossible.

And most importantly—I give you the action plan. The core strategy. The immediate steps to build a resilient AI workflow that doesn’t crumble when half these companies disappear.

❗❗❗You need this information now. Not when your favorite AI tool shuts down with two weeks’ notice. Not when your workflows break because a company pivoted or got acquired. Now, while you can still position yourself strategically.

The ebook is 5000 words of exactly what you need to know, written in the same relatable voice you’re reading right now. Easy to understand. It’s $15—less than one month of any AI subscription you’re probably paying for.

🗣📢 Get the full report on Payhip and learn how to build an AI-assisted writing business that survives the bubble. ‎‧₊˚✧🪩✧˚₊‧

Next week, in Part 3, I’ll show you how to survive, offering actionable takeaways, what happens after the crash—and who emerges on top. But honestly? You need the full picture now. Because the companies you’re relying on don’t have five years. Some of them don’t have five months.

Your move.

---

Maryan Pelland has been making a living as a writer for five decades. She’s watched bubbles inflate and pop, hype cycles come and go, and “revolutionary” tools disappear overnight. She’s still here. Her Pen2Profit newsletter helps writers over 50 build sustainable, profitable writing businesses that survive no matter what Silicon Valley throws at us next.