Michael Burry's AI Short Thesis: Accounting Allegations Against Tech Giants

This analysis is based on the Benzinga opinion piece [2] published on November 12, 2025, which reported Michael Burry’s latest short thesis against AI mega-cap stocks. The Scion Asset Management founder, renowned for predicting the 2008 financial crisis, has disclosed substantial put option positions totaling $1.099 billion against key AI infrastructure companies [1].

The market reaction to Burry’s allegations was immediate and differentiated across targeted stocks. On November 12, 2025, Palantir (PLTR) declined 3.56% to $184.17, while Meta (META) fell 2.88% to $609.01 and Oracle (ORCL) dropped 3.88% to $226.99 [0]. NVIDIA (NVDA), Burry’s primary short target with $187 million in put options, showed remarkable resilience with only a 0.33% decline to $193.80 [0]. The broader tech sector underperformed, with NASDAQ down 0.67% compared to Dow Jones gaining 0.50%, suggesting a rotation away from growth stocks toward value investments [0].

The targeted companies exhibit varying financial profiles that may influence their vulnerability to Burry’s accounting allegations. Meta maintains reasonable valuation metrics with a P/E ratio of 26.19x and strong profitability (30.89% net margin, 43.23% operating margin) [0]. However, Oracle’s elevated P/E of 51.56x combined with recent underperformance (-22.52% over one month) suggests greater vulnerability [0]. Both companies face significant potential earnings adjustments according to Burry’s analysis - Oracle’s profits could be overstated by 27% and Meta’s by 21% by 2028 [1].

Burry’s core thesis centers on alleged accounting irregularities where AI hyperscalers extend asset useful lives beyond realistic 2-3 year product cycles for AI chips and servers [1]. This practice reduces annual depreciation expenses, potentially inflating reported earnings by an estimated $176 billion industry-wide from 2026-2028 [1]. The allegations gain credibility from broader industry concerns, as Bokeh Capital’s Kim Forrest has questioned whether “world needs all the AI data centers that are being built” [3], suggesting potential overcapacity in AI infrastructure.

The convergence of Burry’s short thesis with broader market rotation patterns reveals significant interconnectedness. Recent market data shows growth stocks pulling back 20-30% from highs while value names support broader market performance [1]. This “narrowing leadership” pattern aligns with Burry’s thesis about AI sector vulnerability and suggests systematic reassessment of growth stock valuations.

Burry’s current approach mirrors his 2008 housing market analysis methodology - identifying fundamental accounting irregularities supporting what he perceives as unsustainable booms [1]. However, his historical track record shows occasional early timing, creating uncertainty about immediate market impact. The November 25, 2025 date when Burry promises “more detail” on his analysis [1] represents a critical catalyst point for market participants.

The divergence in institutional responses across targeted stocks reveals nuanced market perception. NVIDIA’s resilience despite being Burry’s largest disclosed short position ($187 million in puts) suggests strong underlying demand or confidence in the company’s fundamentals [0]. Meanwhile, Palantir’s significant decline (-3.56%) reflects both Burry’s substantial short position ($912 million in puts) and potentially greater vulnerability to accounting scrutiny [0].

Michael Burry’s allegations, if validated, could trigger substantial earnings restatements across the AI infrastructure sector. The potential 20-27% earnings adjustments for Meta and Oracle respectively [1] could significantly impact current valuations and investor confidence. potential systematic accounting issues that may warrant regulatory scrutiny and could affect broader AI sector credibility.

• Regulatory investigations by SEC or other bodies into alleged accounting practices
• Potential earnings restatements affecting forward guidance and analyst estimates
• Contagion risk to other AI infrastructure companies not specifically targeted by Burry
• Market psychology effects from a high-profile investor’s public allegations

Key monitoring dates include November 25, 2025, when Burry promises detailed analysis release [1], and upcoming NVIDIA earnings which serve as critical tests for AI momentum [1]. The current market rotation from growth to value may create opportunities in undervalued sectors, while targeted AI stocks could present attractive entry points if allegations prove unfounded or overstated.

Michael Burry’s short thesis against AI mega-cap stocks centers on alleged depreciation accounting manipulation that could overstate earnings by $176 billion from 2026-2028 [1]. His disclosed put option positions total $1.099 billion, with $187 million against NVIDIA and $912 million against Palantir [1]. Targeted stocks showed immediate weakness on November 12, with Oracle declining 3.88%, Meta falling 2.88%, and Palantir dropping 3.56% [0]. The allegations gain context from broader concerns about AI data center overcapacity [3] and ongoing market rotation away from high-growth technology stocks toward value investments [1].

The analysis reveals significant information gaps that require monitoring, including detailed methodology for Burry’s $176 billion calculation, regulatory responses, and company rebuttals. Historical context suggests Burry has identified genuine systemic risks in the past, though timing has occasionally been early [1]. The November 25, 2025 date for Burry’s promised detailed analysis represents a critical information catalyst for market participants [1].