In 2024, Nvidia briefly passed a market capitalization of three trillion dollars, making it one of the most valuable companies in history. Not long after, the launch of DeepSeek R1, a Chinese language model that stunned many with its performance, caused a sharp decline (of 17%) in Nvidia’s stock. This reminded investors just how fragile the current AI boom might be. This back-and-forth of excitement has led many to raise the same question: are we living through another dot-com bubble, or is AI fundamentally different?
At first glance, the similarities are hard to miss. Venture capitalists have poured money into AI at an astonishing pace. In 2024 alone, five American AI startups: OpenAI, Anthropic, Databricks, xAI and Waymo, secured more than thirty billion dollars of investment between them. By mid-2025, AI companies worldwide had already raised one hundred and eighteen billion dollars, with just a handful of firms taking the biggest piece of the pie. OpenAI by itself reportedly drew around forty billion dollars in recent rounds. This sort of concentration is reminiscent of the late 1990s, when anything with “.com” in its name attracted investor cash. The stock market today also shows parallels. The so-called Magnificent Seven dominate U.S. equity performance in much the same way Cisco and Yahoo did during the internet bubble. If investors lose confidence, today’s leaders could see their valuations tested just as harshly.
Yet it would be unfair to stop the comparison there, because there are some important differences. Unlike many dot-com firms, today’s AI companies already have real products that customers pay for. For instance, Microsoft’s Copilot, charges businesses about thirty dollars per user each month to bring AI into everyday programs like Word, Excel and Teams. Analysts believe this could develop into a thirty-billion-dollar annual revenue stream, a vast difference from the empty promises of Pets.com. Meanwhile companies like Nvidia and TSMC are not just riding a wave of hype, they sell the chips and hardware that power AI, gaming, and data centers across industries. Their revenues are diversified and tangible. Beyond the tech giants, AI is also leaving an impact in areas as varied as biotech, where AlphaFold has improved drug discovery, and logistics, where companies are cutting costs with smarter routing. In other words, money is not just chasing an idea, it is chasing working products that already generate returns.
History, however, warns against drawing a simple conclusion. The dot-com crash did wipe out hundreds of companies, but it also laid the foundations for the internet economy we rely on today. Amazon and Google survived the downturn and went on to define the digital era. Another example can be made out of Britain’s railway mania in the 1840s – investors lost fortunes when the bubble burst, but the railways themselves transformed trade and society. Even the cryptocurrency boom of 2017 and 2021 showed how hype can inflate valuations, but some innovations remain in place. AI might follow the same path: lots of money rushing in, a tough crash and then a few strong companies survive and grow.
For investors today, the challenge is to tell facts apart from hype. Almost sixty percent of all AI venture funding is going to just a few headline firms, leaving thousands of smaller startups to fight over what remains. Some of these newcomers are valued in the tens of billions without clear paths to profitability. On the other hand, the companies building the infrastructure: the chips, the cloud platforms, the data centers, seem to be on more solid ground. Another factor to watch is regulation. Policymakers in Washington and Brussels are already debating rules on AI safety, data privacy and competition. Unfavorable laws could slow growth or shift advantage from one region to another, just as privacy regulations reshaped Big Tech in the past decade.
So, is AI the new dot-com bubble? The answer, frustratingly, is somewhere in between. There is no doubt that elements of a bubble are here: valuations that look stretched, a rush of speculative capital and a heavy reliance on a small group of companies. At the same time, AI is already delivering measurable gains in productivity and revenue. The most likely scenario is not a catastrophic collapse, but a shakeout. Many overhyped startups will fall away, while the companies with real products and real business models will continue to thrive.
The lesson for anyone watching the AI boom is the same as it was in 2000 – not every player will survive, but those that do could define the next era of the global economy. The task is figuring out which names are today’s Pets.com, and which are the next Amazon.
– Ignas Vaitekonis
