Are We in an AI Bubble?

Maybe? But what will pop it?

Caroline Swift Holden

Jul 01, 2025

Here’s a question that has been banging in my brain lately.

How do we know if we are in an AI bubble? 🤷🏻‍♀️

Naturally, I asked ChatGPT. And it had a few suggestions:

“How do we know if AI is in a bubble?

Keep an eye on the signals:

💰 Money: Sky-high valuations, VC frenzy, stock spikes
⚙️ Adoption: Real enterprise ROI, not just flashy demos
📈 Impact: Productivity gains vs. job losses and tool fatigue
🔥 Hype: Media buzz, social trends, and too many “AI gurus”

If hype keeps outpacing results, we might be headed for a pop—or at least a sharp correction.”

So let’s take a look at each of these sections and try to determine what our likelihood of a bubble bursting is!

Money Money Money

The global AI market size is $638.23 billion in 2024 and is projected to reach $3.68 trillion by 2034.

OpenAI leads the pack. They just closed a $40B round in March ($30B of which came from Softbank alone 🤯), giving it a $300B valuation. This puts it behind only behind SpaceX ($350B valuation) and TikTok’s parent company Bytedance (also $300B).

Based on valuation alone, OpenAI accounts for nearly half of the entire AI industry’s projected value, an astonishing concentration of hype and capital.

Other startups have raised eye-popping amounts—but none come close to OpenAI.

Anthropic - $18.2B total - $3.5 billion Series E round with a $61.5 billion valuation.
Safe SuperIntelligence - $20B - $30B valuation - No product, just a former OpenAI co founder
Thinking Machines Lab - $2B Seed Round (Ridiculously impressive for a seed round but tiny compared to it’s competition) - $10B valuation - Once again, no product, just a different former OpenAI cofounder
Perplexity - $500M - $14B valuation - numbers so small they look like an aberration

These companies argue that they need this funding to compete against the tech giant: Google, Amazon, Meta (the company formerly known as Facebook), and Microsoft.

Acquisitions and exits are everywhere: heck, I just helped Datavolo get acquihired by Snowflake last November after a year and a month of its existence!

That said, OpenAI (aka half of the AI industry) has some basic structural problems.

It’s still hemorrhaging cash, struggling under a nonprofit-for-profit structure that prevents it from going public—something that could massively improve its financial position. Even more concerning? It still doesn’t have a clear plan for long-term revenue.

It’s what every financial manager you have ever met would describe as a company you probably shouldn’t invest in.

Adoption and Media Hype

AI adoption is large but also extremely varied. ChatGPT is officially the 5th most visited website in the world, narrowly encroaching on Instagram’s numbers, but still nowhere near close Google / Youtube who truly dominate the internet.

Media wise, AI feels like it’s everywhere (at least for me) and definitely made an appearance at the SuperBowl this year but it still doesn’t feel as hyped up as crypto was. 2022’s Superbowl crypto ads were nothing like I had ever seen.

FTX had Larry David, Crypto.com had LeBron James, Coinbase paid how much for a bouncing QR code(???), eToro, Bitbuy: just a wild amount of ad dollars.

AI showed up at the Super Bowl this year, but it still doesn’t feel nearly as hyped as crypto did in 2022.

That said, the one thing that truly feels different to me about AI is not the adoption or the hype: it’s the fear factor.

AI Fear Factor

Crypto felt like an optional gold rush: some people were going after the craze but it certainly wasn’t for everyone.

It wasn’t a threat: simply a question of whether you would have severe FOMO.

AI meanwhile, has an enormous element of fear attached to it, whether you’re an AI fan or not, and has truly captured the public’s imagination.

Many people (including entire industries) are worried about their future employment
Educators are concerned about whether students will actually learn or cheat through life
C Suite executives are spooked by the tech (and intrigued by eliminating employees cough cough) and are demanding that teams “do AI”
Even AI enthusiasts are terrified about what will happen to their loved ones’ careers and lives if they don’t act fast enough

Even governments, terrified of global geopolitical forces, are doing everything they can to remove regulations (via Trump’s Big Beautiful Bill which wants to ban states from regulating AI for 10 years) and accelerate AI growth no matter the costs.

One of my two new interns (Emery Scott) pointed out:

“Humans have reached an age of technological comfort, but the price of that is now we are seeing the next push of tech. This one is terrifying because it isn’t just an awe-inspiring invention, it is something that could outpace us.”

“It makes one wish to slow down for a moment and take a longer look, but then the modern age has never been one to slow itself. It is knowledge that is both a great tool and a dangerous one if kept to the few. AI gives it to the many, so we must keep moving, stay aware of what we can and seek to understand.”

So... are we in a bubble?

💰 Money: 🚨 Yes

⚙️ Adoption: 😬 Mixed but mostly positive?

📈 Impact: 🤷‍♀️ Still early but very promising

🔥 Hype: 🫠 Manageable (for now)

At the moment, I personally don’t think the bubble is going to burst in the immediate future but but I wouldn’t be surprised if a bubble popped in the next two years.

What could cause it? Only time will tell.

But OpenAI’s business structure, if not quickly resolved, spells massive trouble in my book.

Will an AI bubble burst stop the industry? No: it may just deflate for some time and continue to grow over time.

But could a collapse centered on OpenAI devastate the industry potentially beyond recognition?

Possibly.

I’d love to hear from you: Are we headed for a sharp correction, or is AI just getting started?