Situational Awareness LP — хедж-фонд, специализирующийся на искусственном интеллекте, основанный бывшим исследователем OpenAI Леопольдом Ашенбреннером, — seeks to attract additional capital from investors amid a sharp selloff in semiconductor and AI stocks. The move comes at a significant moment: the fund holds the largest publicly disclosed short position against the very companies that drove the semiconductor boom, according to Financial Times. Instead of concentrating its investments in chip-producing companies, the fund has developed a strategy designed to generate returns even as chip manufacturers decline.

Short book against the AI-chip trade

According to its Form 13F filing with the U.S. Securities and Exchange Commission, Situational Awareness LP disclosed a U.S. equity portfolio valued at $5.52 billion as of March 31, 2026. Under SEC rules, institutional investment managers overseeing at least $100 million in qualifying assets must file 13F reports quarterly, giving the public a window — albeit incomplete — into their long U.S. equity positions and certain options positions. The filing showed that the fund's positions were predominantly tied to put options totaling $8.7 billion across companies in the semiconductor and AI sector. Put options give the right to sell a security at a predetermined price and increase in value when the underlying stock falls, making them a common instrument for expressing a bearish view or hedging against declines. These included $2.04 billion in put options on the VanEck Semiconductor ETF and $1.57 billion on Nvidia (NVDA). The fund also allocated more than $1 billion in put options on companies including Oracle (ORCL), Broadcom (AVGO) и Advanced Micro Devices (AMD).

However, the long portfolio tells a completely different story. Instead of opening long positions in chip manufacturers, the fund directed its investments to companies associated with AI infrastructure. Bloom Energy (BE), specializing in hydrogen fuel cells and power generation, became the largest reported long position at $879 million, followed by SanDisk (SNDK) и CoreWeave. The fund also took positions in several bitcoin mining companies, including IREN, Core Scientific, Riot и CleanSpark, as reported by Blockspace. Mining companies have attracted the attention of AI investors because many of them operate or control large-scale, energy-intensive facilities that can be repurposed or co-located for AI computing workloads, positioning them at the intersection of energy and data center demand.

The fund increased several of these investments in mining companies during the quarter, suggesting that power generation and data center infrastructure will benefit from the continued spread of AI, rather than relying solely on the prospects of semiconductor companies.

The cost of one trading session

On July 28, an unverified X account, @LeopoldTracker_, published an estimate according to which Aschenbrenner's portfolio fell by approximately $600 million in one trading session, primarily due to sharp declines in Bloom Energy и SanDisk. Cryptopolitan could not independently confirm this figure.

Leopold Aschenbrenner's fund has lost roughly $600,000,000 in a single session today But he's still up over 1,000% on most of these positions from his entry and his puts are printing: • Bloom Energy, $BE , -$210,000,000 (-14.9%) • SanDisk, $SNDK , -$180,000,000 (-11.0%) • Core… pic.twitter.com/apL862fyLH — Leopold Stock Tracker (@LeopoldTracker_) July 28, 2026

This estimate comes with significant caveats. Form 13F filings capture only certain U.S.-listed equity assets and options held at the end of a given quarter. They do not include cash, derivatives, short positions, foreign securities, or private investments, and therefore provide an incomplete picture of a hedge fund's overall exposure or daily gains and losses.

According to the same tracker, the fund's significant put options may have risen in value as chip company valuations fell, potentially offsetting losses in the long portfolio. While it is impossible to confirm the fund's actual results from public disclosures alone, this dynamic illustrates a hedging approach that fundamentally differs from simply going long on AI stocks.

A billion-dollar bet by an OpenAI alumnus

Aschenbrenner first attracted widespread attention in June 2024 with the publication of his 165-page essay, Situational Awareness: The Decade Ahead. In it, he argued that artificial general intelligence would likely emerge by 2027 and trigger an "intelligence explosion" — a scenario in which AI systems rapidly surpass human capabilities and, in turn, accelerate their own development — leading to massive increases in spending on GPUs, data centers, and electricity, as well as intensifying technological rivalry between the United States and China.

He launched Situational Awareness LP in September 2024 based on this investment thesis. The firm's website describes it as a global equity investor focused on AI as "the dominant driver of global market returns over the next decade." His transition from AI safety research to hedge fund management occurred after his reported departure from OpenAI, where he was involved in policy and safety-related work at the organization.

The thesis resonated with investors. By October 2025, Fortune reported that the fund had surpassed $1.5 billion in assets under management — an outstanding result for a first-time fund manager in his twenties who had previously worked at OpenAI and at the charitable foundation of Sam Bankman-Fried. Supporters characterized him as a pioneer in AI infrastructure investing, while critics noted that the strategy could be too dependent on sustained interest in the sector.

Why raising capital matters as a market signal

The fund's positioning reflects a perspective that is becoming increasingly relevant as AI-related investments evolve. While Aschenbrenner has publicly expressed optimistic views about the long-term trajectory of AI, he has simultaneously made large short bets against semiconductor companies that fueled the sector's rise, while supporting energy suppliers, infrastructure companies, and bitcoin miners.

Raising new capital during a downturn in the semiconductor sector signals conviction that further corrections in chip stocks may lie ahead, even as investment in AI infrastructure continues to grow.

More broadly, the portfolio highlights a challenge that investors may increasingly face: belief in the rise of AI does not guarantee that every segment of the AI supply chain will perform well simultaneously. Capital may shift from purchasing GPUs toward power generation, networking, and data center capacity, expanding the leadership group in the AI sector beyond chip manufacturers.