Is There an AI Bubble or Not?

Industry leaders who believe there’s an AI bubble include Laura Chambers, CEO of Mozilla, who said at the recent Web Summit in Portugal that she sees a typical bubble scenario. 

“The amount of stuff coming out versus the amount of stuff that’s going to [be sustainable] is probably higher than it’s ever been,” Chambers said. “I mean, I can build an app in four hours now. That would have taken me six months to do before. So, there’s a lot of junk being built very, very quickly, and only a part of that will come through. So that’s one piece of the bubble.”

At the same event, David Risher, CEO of Lyft, said AI might be a financial bubble right now because, “no one wants to be left behind,” But he believes it’s a short-term situation, because companies within the bubble are still building tools and learning skills that will be transformational in the long term.

“The data centers and all the model creation, all of that is going to have a long, long life, because it’s transformational,” he explained. “It makes people’s lives easier. It makes people’s lives better.”

Arguments Against a Traditional Bubble

In contrast to those who believe we’re in a bubble, others say there’s no AI bubble because demand for AI far exceeds supply. Firms that supply semiconductor chips, data center infrastructure, and other AI services can’t keep up with the demand. 

These detractors note the situation isn’t a traditional bubble, emphasizing key differences from historical events like the dot-com crash, such as:

Real Revenue and Profitability. The largest companies driving the AI boom — Microsoft, Google, Amazon, and Nvidia, are highly profitable and generate massive cash flows. Unlike many dot-com startups, they’re funding AI investments using their existing profits rather than shaky external financing.

Transformative Utility. The underlying technology has already demonstrated significant real-world utility across countless industries, boosting productivity and creating new services. This differs from the dot-com era, where the  Internet’s business models were often unproven. AI is viewed as a foundational, “general-purpose” technology with decades of adoption ahead.

Massive Infrastructure Investment. The current boom is backed by unprecedented and tangible investment in physical infrastructure, such as data centers, specialized chips (GPUs), and cloud networks. This means the money is building real, lasting assets, not just being spent on marketing or unproven business models.

Widespread Adoption. AI is being adopted at a much faster rate than the Internet was in its early years, with a high percentage of businesses already using some form of AI.

Brad Smith, President of Microsoft, told Fortune that looking at it from a long-term perspective, there’s no bubble. 

“I think that we’ve got years, if not decades, ahead of us for growth,” Smith said. “From a short-term perspective, I’ll only speak for Microsoft; I can’t speak for every company in the industry. We have more demand than supply. That’s the reality of customers, and we have an ongoing pipeline of demand and needs, and we see steady growth, and we’re encouraged by where things are going.”

Some tech leaders and investment analysts say it’s not a bubble because demand for AI far exceeds supply. Ami Badani, head of strategy/CMO at chipmaker ARM, said the company just can’t make chips fast enough.

“Even today, you look at demand and that exceeds supply, and there’s an insatiable amount of demand and appetite for where we need to get to, so I don’t really worry about ‘could folks pull back?’ If you look at GDP growth and where we believe that we need to get to in terms of AI, and these use cases, that’s all limited by power and compute,” Badani said. “So, I don’t worry about it too much.”

What Was the Dot-Com Bubble?

For those of you who aren’t familiar with the beginnings of the Internet and the 1990s dot-com bubble, it was a massive, speculative stock market bubble that caused the excitement and rapid adoption of the Internet when it was relatively new. It peaked in 2000 and ended in 2002.

The Internet was viewed as groundbreaking technology that would change everything. So, investors poured massive amounts of money into any company with a “.com” in its name, believing they would quickly dominate the new digital economy. The focus shifted from traditional business fundamentals, such as revenue, cash flow, and profits, to more abstract metrics like number of website visitors and market share. 

Dot-com companies with no earnings, or even no clear business model, still reached billion-dollar valuations almost overnight. As the frenzy accelerated, startups went public with IPOs at a staggering pace, often seeing their stock prices triple or quadruple on the very first day of trading. This created a sense of “get rich quick” hysteria among investors. 

Investors eventually realized most of the dot-com companies were burning through cash with no strategy for profitability. When funding dried up and confidence waivered, a massive panic sell-off began that caused the Nasdaq index to lose nearly 78% of its value, and thousands of dot-com companies went bankrupt.  

Fortunately, that didn’t kill the technology. Infrastructure such as fiber optic cables and data centers built during the boom remained and paved the way for the next generation of Internet companies that were indeed profitable, such as Amazon and Google. 

The main lesson was that while technology can be revolutionary, its financial success still requires a sound business model and path to profit. As the saying goes, “Those that fail to learn from history, are doomed to repeat it.”