When Will The AI Bubble Burst?
The AI bubble and dot-com bubble look eerily similar. Will the AI bubble burst soon?
My Opinion and Prediction
For those of us old enough to remember the dot-com boom and bust, we are already seeing parallels in the AI boom of today.
In the late 90s, in the eyes of the average person, the internet was a new thing. And it held promise for the future. This was supposed to be the new great thing. A new "dot-com" business model that would revolutionize how business was done.
There was lots of venture money flying around. There were lots of ideas for dot-com start-up companies. And quite frankly, many of them were dumb. One dot-com company's business model was to sell blue-jeans at a loss, then somehow make up the loss by having ads on their website. Needless to say, that particular dot-com did not make it.
In fact, most dot-com start-ups did not make it.
The crash was so great, that the then giant Sun Microsystem was nearly pushed into bankruptcy. During the boom, Sun Microsystems reached a peak market cap of approximately $200 billion. During the bust, Sun servers could be purchased on eBay for 2-5% percent of the price of new systems. There was no reason to buy new Sun hardware. Sun was nearly pushed into bankruptcy and eventally sold to Oracle.
Today we are seeing AI startups that are built on dumb ideas and others that are poorly implemented. One example, I tried to find an AI site to give me an AI summary of a video. From the search engine, I clicked on 3-4 websites. Some wanted me to register and create an account. Some did not seem to work at all. Around the 4th or 5th result I finally found one that would give me what I wanted.
Most AI startups are not profitable, as many are struggling with high operational costs and significant investments without corresponding revenue. Even the world-famous OpenAI is not expected to be profitable until 2030.
What is particularly interesting is that on forums, people are bringing up FuckedCompany.com in 2026 when talking about the current AI/startup bubble. Recent discussions about AI-company failures explicitly invoke that site with nostalga as something people wish existed today.
How much longer do we have in the AI boom before we see the AI bust? It is my opinion that if we follow the same clock cycle of the dot-com boom/bust, we are two to three years away from the bust.
The good news is, after the crash, we will be able to buy RAM again.