Are analysts comparing this to the dot-com crash?

Some are. Brokerage BTIG warned that certain signals — heavy use of leveraged ETFs and options by retail traders, which amplifies swings in chip stocks — resemble the 2000 dot-com bubble.

Did Micron's own business actually weaken?

No. Analysts describe Micron as a "strong company, weak tape": it still runs high margins and strong cash flow, and demand for AI memory chips remains above what factories can produce.

Chip Stocks Slide on AI Bubble Fears

A widening selloff in semiconductor shares deepened this week after AMD's latest earnings failed to satisfy investors, dragging down Nvidia, Micron and other chipmakers tied to the artificial intelligence boom. AMD beat sales and profit forecasts when it reported on Aug. 4, yet its stock still fell about 8%, as traders fixed on a dip in profit margins and asked how long the spending surge behind these companies can last.

The company posted quarterly revenue of $11.54 billion, up 52% from a year earlier, and adjusted earnings of $1.66 a share, both above Wall Street estimates. It also raised guidance for the current quarter to about $13 billion. But its gross margin came in at 54%, short of the 56% analysts wanted, which AMD blamed on the cost of ramping up its new Helios AI systems.

The reaction fit a familiar pattern. AMD shares had climbed roughly 132% this year before the report, leaving little room for anything less than a blowout. Chip stocks as a group have run hard and are now pulling back: the Philadelphia Semiconductor Index has fallen more than 20% from its June record.

The pressure began weeks earlier. In late July, more than $1.3 trillion in market value was wiped from the world's largest chip companies in a handful of trading sessions. Nvidia lost about $238 billion, while memory makers SK Hynix, Samsung and Micron shed $176 billion, $173 billion and $113 billion. On the worst day, Micron and SK Hynix each dropped close to 10%.

Beyond the daily swings, the deeper question is whether the money pouring into AI data centers will pay off. "Investors are reassessing whether near-term revenues can justify unprecedented AI spending levels," said Charlie Dai, vice president and principal analyst at Forrester, who called the drop "less about weakening AI demand and more about a repricing of expectations" after a strong rally. UBS estimates spending by the largest cloud companies will rise 76% this year, to about $673 billion.

Not everyone sees a bubble bursting. Stacy Rasgon, a senior analyst at Bernstein who rates AMD a buy with a $600 price target, said he "saw nothing wrong" with the results. AMD chief executive Lisa Su said demand for the company's accelerators and processors is "growing well above our prior expectations" and predicted data-center sales would double in 2027.

The selling looks more like a repricing than a retreat. Memory shortages tied to AI are still expected to last into 2028, and banks such as J.P. Morgan and Bank of America frame the pullback as a buying chance. What investors will watch now is whether the big cloud buyers keep funding the boom, and whether fresh competition - including reports that China's DeepSeek is designing its own AI chip - starts to erode the pricing power that made these stocks so valuable. Until AI spending translates more clearly into profit, chip shares are likely to stay volatile.