67% Monthly Crash: Situational Awareness Fund Liquidates on Chip Overexposure

Woofun AI reports that Situational Awareness, the AI hedge fund managed by Leopold Aschenbrenner, underwent liquidation in July following a catastrophic 67% decline in net value within a single month.

The magnitude of this erosion underscores a sharp pivot in investment trends within the AI sector. Profit margins, once expansive for beneficiaries of AI data center growth, are narrowing significantly. The fund's heavy concentration in memory stocks, which experienced precipitous drops, illustrates how rapidly favorable conditions can reverse into severe losses.

Regulatory filings submitted to the SEC on August 14 reveal the structural vulnerabilities present before the collapse. The 13F-HR form, detailing holdings as of June 30, showed that SanDisk and Micron collectively comprised approximately 50% of the fund's reported stock assets. This extreme concentration indicated that the portfolio was dangerously exposed to a single link in the AI supply chain, leaving little room for error when sector dynamics shifted.

Earlier portfolio data from the March 31 quarterly filing further illustrates the scale of the fund's aggressive positioning. At that time, Situational Awareness held U.S. stocks valued at $5.52 billion and $8.7 billion in put options related to chip companies. Its largest long position was Bloom Energy, with an $879 million investment, followed by significant stakes in SanDisk and CoreWeave. The total securities value stood at $13.68 billion, a figure that reflects economic exposure rather than total fund assets.

These positions were initially aligned with robust market demand drivers and favorable forecasts. Counterpoint Research noted that as AI workloads transitioned from training to inference, enterprise SSDs accounted for 48% of total NAND shipments by the second quarter of 2026. TrendForce predicted that NAND Flash revenue would increase by 10% to 15% quarter-on-quarter in the third quarter of 2026, while DRAM prices were expected to rise by 13% to 18%.

Micron's performance appeared to validate these bullish assumptions, reporting record results for its third fiscal quarter with revenue reaching $41.46 billion, a stark contrast to $9.3 billion in the same period last year.

However, despite recognizing these trends, the fund could not escape the liquidity crisis.

Woofun AI data shows that after liquidity dried up, the fund was forced to reduce public holdings to lower leverage, resulting in an unaudited 67% loss in July, though the return rate from the beginning of the year remained close to 80%.

Investor communications and unverified estimates add further context to the turmoil. Aschenbrenner acknowledged the risks, stating, "We embrace volatility, but it must never put the fund at risk," and noted steps taken "to fight again in the future." On July 28, an anonymous X account @LeopoldTracker_ posted an estimate suggesting the portfolio lost around $600 million due to declines in Bloom Energy and SanDisk.

However, this figure remains unverified, as 13F filings do not include cash, short positions, or personal investments.

Despite the liquidation, Aschenbrenner maintains optimism about the broader AI landscape. His firm invested $400 million in Source Foundry, a secretive semiconductor equipment startup valued at nearly $5 billion and focused on lithography technology, a bottleneck dominated by ASML. This aligns with macro projections from Goldman Sachs, which estimates that spending in computing, data centers, and power sectors will reach nearly $7.6 trillion between 2026 and 2031.

Additionally, SIA and Deloitte state that over 95% of the value of AI server racks comes from semiconductors, with data center infrastructure investments potentially reaching $4 trillion by 2028.

The failure of Situational Awareness highlights that belief in AI growth does not guarantee survival for every link in its supply chain. Highly leveraged, concentrated positions remain vulnerable to liquidity crises, regardless of accurate demand assessments. This case serves as a stark reminder that financing methods can eliminate investors even when their fundamental thesis is correct.