Transcription
Is artificial intelligence the new dotcom bubble? It is a question echoing across boardrooms, trading floors, and tech conferences. And for a good reason.
The excitement around AI today feels almost identical to the frenzy that swept the world during the late 1990s internet boom. Investors, governments, and everyday users are all captivated by the promise of this technology. And yet beneath the enthusiasm lies a growing sense of deja vu. We've seen the movie before and we know how the first act usually ends. But the bigger question is whether the AI revolution is genuinely different or whether history is preparing to repeat itself.
The numbers alone are dizzying. OpenAI has raised almost $63 billion since 2015. Entropic, founded only in 2021, has pulled in around $27 billion in just four years. And to put that into context, SpaceX, one of the most ambitious engineering projects in human history, spent two decades raising $12 billion. And its $400 billion valuation has now been overtaken by OpenAI's $500 billion valuation. These are not, of course, normal figures. These are the hallmarks of a market that believes the future has already been decided and that missing the next big thing is actually worse than placing a reckless bet.
That belief hit an almost surreal moment in early September when Oracle reported earnings that actually missed expectations. Ordinarily, that would drag a company's stock down. But instead, it surged by 36% in a single day after executives predicted all of a sudden that AI-powered cloud services would generate $144 billion in revenue by 2030. So, one mention of AI added a quarter trillion dollar to its market cap overnight. Even Sam Altman, the face of the boom, has warned, "Are investors overexcited about AI?" My opinion is yes. He said, "When the people at the center of the excitement start sounding alarms, you know something unusual is happening."
And the parallels to previous bubbles don't really end there. During the dotcom era, companies with barely a prototype or no product at all achieved sky-high valuations simply for adding ".com" to their names. Do you remember that? It happened again during the 2017 crypto mania when businesses slapped blockchain on their branding and watched their stock prices explode, often without implementing any blockchain technology at all. Today, AI has become the latest magic keyword. Attach it to a pitch deck and all of a sudden, investors stop asking hard questions and the money just flows.
This dynamic is fueled by a major knowledge gap. Most investors, journalists, and consumers don't really understand AI any better than the public understood the internet back in 1998. That makes it easier for flashy demos and sleek prototypes to mask flaws, limitations, and sometimes outright illusions. Viral clips of humanoid robots mixing cocktails and chatting with humans rarely mention that some of these robots actually are controlled by remote operators hiding backstage. Similarly, many popular new AI apps aren't building new technology at all. They're simply wrappers that sit on top of existing models like ChatGPT. Grammarly, for example, Copy AI, and dozens of others fall into this category. They're useful, yes, but they're not exactly technological breakthroughs. They exist because someone took an existing AI model and built a friendly interface for everyday users. That's all there is to it.
With so much hype, it is no wonder that comparisons to the dotcom bubble keep surfacing. We keep hearing it over and over again. But drawing a perfect parallel would be a mistake in my opinion. This time, the foundations beneath the hype are very different and it could actually fundamentally reshape what a bubble even looks like. In the 1990s, many tech startups were run by students and those who were just who perceived it as a hobby who experimented out of garages. Well, today the biggest AI models are backed by vast corporate empires such as Microsoft, Google, Amazon, companies with global distribution networks, virtually unlimited capital, and entrenched influence over both governments and everyday life. In this case, mass adoption isn't a dream anymore. It is essentially guaranteed.
Here's what's very important and what's different this time around. Governments are no longer neutral bystanders in this process. They see AI as a strategic technology that is tied to geopolitical power. And that means contracts, subsidies, tax breaks, and preferential treatment. And that, of course, becomes an artificial boost, if you will, that can keep companies alive regardless of market forces, regardless of true demand. In other words, when the state wants you to win, the playing field ceases to be fair. It becomes engineered. And that alone makes the AI boom fundamentally different from the dotcom era or from any other bubble that we've experienced before.
Still, even with these advantages, markets won't stay irrational forever. It's impossible. Just like the internet, AI is here to stay. But the industry around it is absolutely in bubble territory. Investors are throwing money at flimsy products. Companies are slapping AI onto their branding to look innovative. And sooner or later, the market will correct. Many of the apps and startups that look unstoppable today will vanish. There's no question about that. But unlike past bubbles, the AI collapse won't wipe out everybody equally because of the enormous corporate and government backing behind the largest models. The winners will be few, but they will be unbelievably powerful.
What separates those winners from the losers will be adaptability. For example, look at Netflix. It began as a DVD-by-mail service and then it pivoted to streaming at the right moment and it became a global powerhouse of the internet age. Today's tech giants are already trying to make their own pivots for the AI era. For example, Google's AI Overview feature is reshaping the internet's traffic patterns by providing direct answers instead of sending users to websites. Have you noticed that? Let me know in the comments below. And because of that, some publishers say that their click-through rates have fallen nearly 90% in certain categories. These companies are reshaping Facebook and Instagram and other social platforms with AI-generated content, AI influencers, and AI-driven tools. And yet, even they admit that they don't fully know whether this will succeed. The underlying lesson is very clear: Adapt or get crushed.
There is, however, another possibility. Maybe the hype isn't misplaced at all. AI is improving in real time. It is improving as we speak. Today's limitations could actually evaporate faster than anybody anticipates. This is a new area and if that happens, then the astronomical valuations that we see today and the frantic investment rush might actually turn out to be perfectly rational. It is possible, and instead of a bubble waiting to burst, we might be standing on the edge of one of the greatest economic expansions in modern history. Although, if you ask me, the first scenario appears to be far more likely than an enormous economic breakthrough in this case.
So, is AI the new dotcom bubble? The answer is both yes and no, in my opinion. Yes, the hype is enormous and yes, many investors are actually blindly chasing it. But no, the structure of the AI industry is not the same as in past bubbles. And the forces behind it, such as corporate power, geopolitical competition, and a genuinely transformative technology. These forces are far stronger than anything we saw before. The bubble may pop, but the technology is definitely here to stay.
Let me know what your thoughts are on the AI bubble. What do you expect? I would love to hear from you in the comments below. As always, thank you for watching. And if you'd like more content, join me on Substack and Patreon. Support my work. I appreciate every single one of you who has become a paid subscriber. Thank you so much. Like, subscribe, and share. And I will see you back here tomorrow. Bye for now.