Michael Burry Just Shorted Micron Stock. Micron Just Printed Records.

Michael Burry disclosed a direct short in Micron Technology at $1,051.87 on July 1, 2026, according to a Substack post first reported by Invezz, arguing the chipmaker’s roughly 260% year-to-date rally has been driven by investor psychology rather than underlying business strength. The trade sits next to disclosed shorts in Nvidia, Applied Materials, and the iShares Semiconductor ETF, with Burry stating AI-related chip names could face a 30% correction.

The move lands against Micron’s own fiscal third-quarter 2026 results, filed June 24, 2026, which showed record revenue, the highest gross margin in company history, and the largest quarterly free cash flow of any memory cycle on the public record. Two sets of facts, both fully sourced, are pointing in opposite directions, and the trade is a wager on which one bends first.

The Trade, as Burry Wrote It

In his Substack post, Burry made the case that Micron’s rally reflects crowd behavior more than durable earnings power. He wrote that the stock is being driven by “fear of missing out, greater fool theory, [and] public commitment bias,” the Invezz summary of the post records. Burry pegged Micron’s median return on invested capital at 4% and median return on equity at 7%, calling both figures “frankly terrible,” and added that “one quarter in every three, Micron is a destroyer of capital,” a pattern he tied to decades of inconsistent free cash flow.

His historical lens is the load-bearing part of the trade. Micron “defines cyclical like no other,” Burry wrote, noting the company has recorded 34 drawdowns of more than 30% over the past 42 years. He pointed out that shares are now trading further above their 200-day moving average than at any point since 1984, “not even during the dot-com peak.” Asked why he used a direct short rather than puts, Burry said options were expensive: “the puts seemed expensive,” adding he “will look to add puts should the stock settle down and bring volatility down.”

What Micron Actually Just Printed

The fiscal third quarter ended May 28, 2026, and the numbers do not match the “destroyer of capital” framing. For the period, Micron posted revenue of $41.46 billion, up from $9.30 billion in the same quarter a year earlier, according to the company’s June 24, 2026 earnings release. Gross margin reached 84.6%, against 37.7% a year earlier. Operating margin came in at 80.4%, against 23.3%. Net income for the quarter was $28.24 billion, against $1.885 billion a year earlier.

Cash generation tells a similar story. Operating cash flow reached $25.39 billion in the quarter, against $4.61 billion a year earlier, and adjusted free cash flow came in at $18.3 billion. Micron ended the quarter with $30.2 billion in cash, marketable investments, and restricted cash, with long-term debt down to $5.14 billion from $14.02 billion as of the prior fiscal year-end. The company guided fiscal Q4 2026 to revenue of $50.0 billion, plus or minus $1.0 billion, with a non-GAAP gross margin of approximately 86% and non-GAAP diluted EPS of $31.00, plus or minus $1.00.

The Memory Cycle, and What Changes This Time

Memory has historically been one of the most cyclical industries in technology. Demand rises, manufacturers add capacity, supply catches up, prices fall, margins compress, and the cycle repeats. Burry is betting this is the moment the script finally executes, and a Budgy App analysis of Micron’s HBM positioning explains the other side: HBM3E capacity per Nvidia GPU went from 141 GB in the H200 to 288 GB in Blackwell Ultra, and Micron is in high-volume production on HBM4 for Nvidia’s Vera Rubin platform, a generation the wider memory industry has not yet matched at scale.

That generational step-up shows up in Micron’s quarterly business units. In fiscal Q3 2026, the Cloud Memory Business Unit posted $13.77 billion in revenue with a 78% operating margin, against $3.39 billion and a 46% margin a year earlier, per the same earnings release. The Core Data Center Business Unit posted $11.52 billion in revenue and an 83% operating margin, against $1.53 billion and a 20% margin a year earlier. The structural argument is that high-bandwidth memory carries a margin premium commodity DRAM does not, because the die stacking is hard to replicate. The cyclical argument is that premium eventually commodifies, and memory has a long history of punishing the suppliers that built too much.

On the June 24, 2026 earnings call, Micron Chief Business Officer Sumit Sadana said customer demand for memory chips remains “well above our ability to supply” across nearly every product category through 2028, and CFO Mark Murphy said free cash flow has grown to levels not seen in most of the company’s history, per the TheStreet earnings recap. Burry’s thesis survives only if those 2028 supply commitments break earlier than the customer roster expects.

The Rest of Burry’s Chip Book

The Micron short is one line in a broader bearish semiconductor position. According to Invezz, Burry now holds disclosed shorts in Nvidia (NVDA), Applied Materials (AMAT), and the iShares Semiconductor ETF (SOXX), and has said AI-related chip names could see a 30% correction. In a separate June 30, 2026 Substack post covered by the same outlet, Burry said plans by Samsung Electronics and SK Hynix to invest more than $500 billion in a new semiconductor hub represented “the beginning of the end” for the rally.

The market backdrop heading into Burry’s short disclosure was already weakening. Micron shares fell about 5% on Thursday and nearly 11% on Wednesday after reaching a high of $1,255 following its June 25, 2026 earnings reaction, with some market participants linking the slide to reports Meta is considering selling excess cloud capacity, while a separate report indicated Apple is seeking additional memory supply from China, per the Invezz write-up. Swissquote senior analyst Ipek Ozkardeskaya commented that “China makes around 15% of Apple’s sales and other companies could follow these steps as they also see their profits being squeezed by an unreasonable jump in memory chip prices.” Apple is not among Burry’s disclosed chip shorts.

Timing Is the Whole Bet

Burry is not arguing that Micron’s recent quarter was bad. He is arguing that the price has run too far, too fast, and that the stock’s own historical pattern of sharp reversals makes a pullback more likely than not. Micron’s track record of 34 drawdowns of more than 30% over four decades is real, and the stock is trading at the most extreme extension above its 200-day moving average on record.

Whether that timing proves right depends on how long the current memory shortage and the AI spending cycle last. Invezz summarizes Burry’s own framing of the position he also increased on July 1, adding to long holdings in PayPal, Sprouts Farmers Market, Zoetis, Fannie Mae, and Freddie Mac, a roster that suggests the chip short is a single line in a portfolio, not a directional bet. His summary line: “Yesterday I shorted one stock even though it was down a good amount because I think I have a pretty good idea how this resolves. I also added to five positions. This time may be different, but not nearly different enough.”

Disclaimer: This article is for informational purposes only and is not investment advice. Short positions, semiconductor stocks, and memory names in particular carry significant cyclical and concentration risk. Figures cited are accurate as of publication and sourced to company filings or named-report coverage. Readers should consult a qualified financial professional before making investment decisions.