Is there a bubble in AI? The answer is definitely yes.

Looking at recent storage and semiconductor stocks, SanDisk has risen 410% in four months, Micron doubled, SK Hynix increased by 135%.
The market is using real money to prove that AI demand is genuine.
But genuine demand doesn't mean the prices are reasonable.
The seven tech giants are spending over $600 billion this year, burning money at a rapid pace, and infrastructure is continuously improving, but there seem to be very few applications that truly change ordinary people's lives.
The pace of capital expenditure far exceeds the speed of application deployment.
Before the dot-com bubble burst in 2000, internet demand was also real.
The internet did indeed change the world, but chip stocks fell 80%, and it took a long time to return to their highs.
The technology was correct, but the pricing was wrong.
Bubbles can last a long time, long enough for everyone to think there is no bubble, and then they burst.
So the current question isn't whether there is a bubble, but when it will burst and how.
I believe no one knows the answer to that.