Even Sam Altman admits investors are "overexcited" about AI. Here's why history might be repeating itself—and what it means for your money.
When the CEO of a company valued at half a trillion dollars tells you that investors are "overexcited" about his own industry, you should probably pay attention.
In August 2025, OpenAI's Sam Altman did exactly that. "Are we in a phase where investors as a whole are overexcited about AI?" he asked rhetorically. "My opinion is yes."
He's not alone in thinking this. Last week, the International Monetary Fund compared the current AI investment boom to the dot-com bubble that devastated the tech sector in 2000. Jeff Bezos called it an "industrial bubble." Mark Zuckerberg acknowledged the parallels. And now, with tech companies pledging a staggering $400 billion in AI infrastructure spending for 2025 alone, the question isn't whether we're in a bubble—it's whether we're about to watch it burst.
The Numbers That Should Terrify You
Let's start with the most shocking statistic: Global corporate AI investment reached $252.3 billion in 2024, growing thirteenfold since 2014. To put that in perspective, tech companies are now collectively funding the equivalent of a new Apollo moon landing program every ten months.
But here's where it gets interesting—and alarming.
The "Buffett Indicator," which measures the stock market's value against GDP, currently sits at 217%. That's higher than the 210% peak during the dot-com era, and significantly above the 140% level that preceded the 2008 financial crisis. Some economists now estimate that the AI bubble is 17 times larger than the dot-com frenzy.
Meanwhile, Amazon, Google, Meta, and Microsoft are spending an average of $36 billion each per quarter on AI infrastructure—compared to just $2 billion for a typical S&P 500 company. And get this: nearly 40% of US GDP growth in the first half of 2025 came from AI-related tech spending. Without it, the American economy would have grown at just 0.1% annually—essentially flatlining.
The Company Worth $500 Billion That Doesn't Make Money
Perhaps no company better illustrates the bubble dynamics than OpenAI itself. The company's valuation has nearly doubled from $300 billion to $500 billion in less than a year, making it the most valuable privately held company on Earth.
The problem? OpenAI is hemorrhaging money.
The company recently committed to spending $300 billion with Oracle over five years—that's $60 billion annually. Yet its projected revenues for 2025 are only $13 billion. That means OpenAI is spending more than four times what it earns, with no clear path to profitability in sight.
And they're not outliers. MIT's 2025 "State of AI in Business" report found that despite $30-40 billion invested in enterprise AI, 95% of generative AI pilots at companies are failing to deliver measurable returns on investment.
Sound familiar? It should.
Déjà Vu: When the Internet Was Going to Change Everything (Too Fast)
Between 1995 and 2000, the NASDAQ composite index rose 600%. Internet companies with no revenue, no profits, and often no viable business plan commanded billion-dollar valuations. Investors threw money at anything with a ".com" in its name.
Then reality hit.
On March 10, 2000, the NASDAQ peaked at 5,048.62. By October 2002, it had crashed 78%, wiping out trillions of dollars. The index wouldn't recover to its 2000 peak for 15 years. Thousands of companies went bankrupt. Millions lost their jobs.
But here's what most people forget: the Internet didn't fail. The valuations did.
The same telecommunications companies that fueled the dot-com boom laid over 80 million miles of fiber optic cables across the United States. Four years after the crash, 85-95% of that fiber remained unused—earning the grim nickname "dark fiber." Corning, the world's largest optical fiber producer, saw its stock plummet from nearly $100 to about $1.
The infrastructure was real. The technology was transformative. But demand couldn't justify the massive upfront investment—at least not in the short term.
The Three Ways This AI Bubble Could Pop
Yale researchers have identified three distinct scenarios for how the AI bubble might burst:
1. The Circular Investment Trap
Right now, a small group of companies—OpenAI, Nvidia, Microsoft, Google, AMD, and a few others—are essentially investing in each other in an increasingly tangled web of deals.
Nvidia is investing $100 billion in OpenAI. OpenAI is taking a 10% stake in AMD. Microsoft is both a major OpenAI shareholder and a customer of AI cloud company CoreWeave, which Nvidia also owns a stake in. And Microsoft accounts for almost 20% of Nvidia's annual revenue.
When everyone's success depends on everyone else continuing to spend, what happens when one link in the chain breaks?
2. The "Dark Data Center" Problem
Just as the 1990s left behind miles of unused fiber optic cables, today's AI boom could create massive "stranded assets"—data centers, chips, and infrastructure that sit idle because demand never materialized.
Oracle recently revealed it expects to "lose considerable sums of money" on its data center rentals, primarily to OpenAI, already incurring $100 million in losses in just one quarter. The company's shares initially soared 40% on the OpenAI deal announcement, adding nearly one-third of a trillion dollars to its market cap. But if OpenAI can't generate enough revenue to keep paying, that value evaporates.
3. The Federal Reserve Reality Check
The dot-com bubble didn't burst because of a single catastrophic event. It burst because the Federal Reserve raised interest rates six times between 1999 and 2000, making speculative tech investments less attractive compared to safer bonds.
Sound familiar? With inflation concerns still lingering and the economy heavily dependent on AI spending, any significant interest rate hikes could trigger the same reassessment of risk that cratered the dot-com market.
But Wait—Isn't AI Different?
The tech defenders have a point: AI is already showing real-world applications in ways the early Internet couldn't match.
Unlike 1999, when most dot-coms were pure speculation, AI is being integrated into healthcare diagnostics, manufacturing optimization, drug discovery, and enterprise automation. These aren't hypothetical future benefits—they're happening now.
Goldman Sachs estimates AI has already added about $160 billion to the US economy since 2022, or roughly 0.7% of GDP. And unlike the highly leveraged housing bubble of 2008, today's AI investments aren't built on debt—they're funded by cash-rich tech giants.
As the IMF's chief economist noted, "if there is a market correction, some shareholders, some equity holders, may lose out. But it doesn't necessarily transmit to the broader financial system."
In other words, when (not if) this bubble pops, it probably won't crater the global economy like 2008 did. But that's cold comfort if your retirement portfolio is heavily invested in tech stocks.
The Uncomfortable Truth
Here's what we know for certain:
The technology is real. AI will transform industries, boost productivity, and change how we work and live—just as the Internet did.
The valuations are insane. Companies valued at hundreds of billions despite losing money aren't sustainable in the long term.
History rhymes. Every major technology boom—railroads, radio, the Internet—has followed the same pattern: initial hype, massive overinvestment, painful correction, then long-term transformation.
The question isn't whether AI will change the world. It will. The question is whether you want to be holding the bag when reality catches up to the hype.
What This Means for You
If you're an investor, the lesson from the dot-com era is clear: be selective. Amazon and eBay survived the crash and thrived. Pets.com and Webvan didn't. The companies with actual revenue, sustainable business models, and genuine competitive advantages will weather the storm.
If you're a worker in tech, brace yourself. The dot-com crash triggered mass layoffs and took four years for employment to recover. Even if your company survives, the correction will be painful.
And if you're just watching from the sidelines? Understand that this AI boom is now a significant driver of the entire US economy. When it slows—and it will—we're all going to feel it.
The dot-com bubble taught us that transformative technologies don't follow straight lines to success. They follow hype cycles, boom-bust patterns, and painful corrections before they truly change the world.
The only question now is: have we learned anything from history, or are we destined to repeat it?
Frequently Asked Questions
Is the AI bubble going to crash in 2025?
While no one can predict exact timing, several warning signs suggest a correction could happen soon. The Buffett Indicator is at historically high levels (217%), and 95% of enterprise AI projects are failing to deliver ROI. However, unlike the dot-com crash, this won't necessarily happen overnight—it could be a gradual deflation rather than a catastrophic burst.
Should I sell my tech stocks now?
Not all tech stocks are equal. Companies with actual revenue, diverse business models, and profitability (like Microsoft, Google, Apple) are much safer than pure-play AI startups burning cash. The dot-com crash taught us that Amazon and eBay survived while Pets.com didn't. Focus on fundamentals, not hype.
How is the AI bubble different from the dot-com bubble?
Three key differences: (1) AI is already showing real-world applications and generating revenue, unlike many 1999 dot-coms; (2) Today's investments are funded by cash-rich companies, not massive debt; (3) The technology itself is more mature and proven. However, the valuation dynamics and hype cycles are remarkably similar.
What happened during the dot-com crash?
The NASDAQ fell 78% from March 2000 to October 2002, wiping out trillions in wealth. Thousands of companies went bankrupt, millions lost jobs, and it took 15 years for the index to recover. However, the Internet itself continued to grow and eventually transformed society—just not as quickly as investors hoped.
Will AI companies survive if the bubble bursts?
The strongest ones will. Companies with sustainable business models, actual customers paying for products, and diversified revenue streams will weather the storm. But startups valued at billions despite losing money annually will face severe pressure when investor sentiment shifts and funding dries up.
How much money has been invested in AI?
Global corporate AI investment reached $252.3 billion in 2024, growing thirteenfold since 2014. For 2025, the top four tech companies (Amazon, Google, Meta, Microsoft) are pledging approximately $400 billion in AI infrastructure spending combined. OpenAI alone has committed to $300 billion in spending with Oracle over five years.
What are "dark data centers"?
This refers to the potential for massive AI infrastructure—data centers, chips, servers—to sit idle or underutilized if demand doesn't materialize as expected. It's similar to the "dark fiber" problem from the dot-com era, when 85-95% of fiber optic cables laid in the 1990s remained unused for years after the crash.
Is Sam Altman really worried about an AI bubble?
Yes. In August 2025, Altman publicly stated that investors are "overexcited" about AI as a whole. However, he maintains that OpenAI specifically will deliver transformative value. This nuanced position—acknowledging bubble dynamics while defending his own company—mirrors what many tech CEOs said during the dot-com era.
What should regular people do about the AI bubble?
Stay informed but don't panic. If you're investing, diversify and focus on fundamentals over hype. If you work in tech, build transferable skills and maintain an emergency fund. If you're a business owner, be cautious about massive AI investments until ROI is proven. The technology will transform the world—just maybe not on the timeline Wall Street expects.
Could this crash the entire economy like 2008?
Unlikely. According to the IMF, today's AI investments are funded by equity, not debt leverage like the 2008 housing bubble. If there's a correction, shareholders will lose money, but it shouldn't trigger a cascading financial crisis. However, since AI spending now represents 40% of US GDP growth, a slowdown would definitely impact the broader economy.
What do you think? Are we in an AI bubble, or is this time really different? Drop your thoughts in the comments below.
Post a Comment