Burry's Fresh Nvidia and Micron Warning: Smart-Money Hedge or Late-Game Short Squeeze Bait?

Burry's Latest Moves Look Like Downside Insurance

In his latest update, Burry added Nvidia puts expiring Dec. 18, 2026 with strikes in the low $100s and increased bearish exposure to Micron, SOXX, and QQQ puts. That does not mean he is declaring an AI top today. It does mean he is paying for protection while the semiconductor trade remains crowded.

What the filings actually show

The structure matters. Burry chose dated Nvidia puts rather than relying only on portfolio reshuffles, which looks more like downside insurance than an all-out exit call. With strikes in the low $100s, he is pricing protection into next year, not just the next few weeks.

On Micron, he first disclosed a short near $1,051.87 after saying puts "seemed expensive," then later added to the Micron short around $880. He also expanded a SOXX short near $506. Taken together, the filings suggest selective bearish exposure rather than a blanket rejection of the sector.

The main risk for late shorts is still timing. If semiconductor momentum holds, these trades can look expensive and premature for a while longer.

Burry's Core Bet Is a Cyclical Repricing, Not Just a Bad Quarter

A 30% correction is a cycle call

Burry said AI-linked chip stocks may be due for a 30% correction. That frames the risk as a multiple-and-cycle issue, not just an operational miss. His Micron thesis rests on the same logic: when a cyclical stock runs hard, even a strong quarter may only delay the reset.

According to the reporting, Micron had risen 242% year to date and was valued at about $1.17 trillion at one point. Burry also pointed to cycle history, noting 34 drawdowns of more than 30% over 42 years. The argument is that extreme extensions can give way to sharp re-rating even while AI demand still looks strong on the surface.

Aschenbrenner's filing shows the same fault line

Burry is not the only investor buying downside exposure. Leopold Aschenbrenner's 13F showed he opened about $8.45 billion in new puts against tech and semis. His disclosed put positions included SMH, Nvidia, Oracle, Broadcom, AMD, Micron, and Taiwan Semiconductor. The size of that exposure suggests some investors see too much optimism built into the sector, even if it does not prove a near-term crash.

Why the debate matters now

Bulls can argue that AI spending is structural and that old cycle charts do not apply in the same way. Bears can argue that a supernormal demand phase often sounds safest right before expectations normalize.

That is why the filing matters less as a trading signal than as a watchpoint. If customer budgets, supplier guidance, or financing conditions weaken, multiple compression can hit before fundamentals fully break.

A Bearish Filing Is a Signal, Not a Trading Order

A bearish 13F can be useful, but it is not proof that a stock must fall.

Separate "bearish" from "right"

Scion's latest filing showed disclosed shorts in NVIDIA around $186.5 million and Palantir around $912 million. That is useful as an attention signal, especially when a known contrarian investor puts real capital behind a thesis. But it is still only one data point.

The limitation is mechanical. A 13F is a snapshot, reported retrospectively, and for options it reflects notional values rather than premium paid or actual P&L. In plain English, the filing shows exposure size, not timing, confidence, or exit plan.

That is why Nvidia and Micron should not be traded bearishly just because Burry is bearish. Investors should wait for evidence that positioning risk is turning into operating risk.

What Would Confirm the Bust-and What Would Confirm the Trap

This is a late-2026 monitoring window, not necessarily a next-week trade. The key date is Burry's Nvidia puts expiring Dec. 18, 2026; after that, the filing becomes history rather than an active setup.

What to watch if the bearish thesis is right

• AI customers or suppliers begin emphasizing budget prioritization instead of open-ended demand, with hyperscaler capex commentary becoming more measured.
• Management teams find it harder to defend rich valuations because execution or demand language weakens.
• Investors start questioning whether HBM demand alone is enough to keep supporting premium multiples.
• The market rewards results more than narrative, making guidance quality more important.

What to watch if the bearish thesis is wrong

• Nvidia and Micron keep absorbing capital while customers still find ways to fund the buildout.
• Companies keep beating and raising expectations, showing that AI demand still has real cash behind it.
• Skepticism stays narrative-driven rather than operating-driven, increasing the risk of a squeeze.

Until that evidence arrives, this looks more like a watchlist call than a trigger.