His firm, Situational Awareness LP, recently disclosed a massive portfolio shift, revealing over $7 billion in put-option positions against major semiconductor and AI-infrastructure leaders Nvidia, AMD, Oracle and others.
Aschenbrenner is of the view that the biggest bottleneck for AI to develop will be physical infrastructure, particularly electricity and computing power.
The Big Short Positions
In his latest 13-F disclosures, Aschenbrenner’s firm Situational Awareness disclosed a $2.04 billion short position on the VanEck Semiconductor ETF, a gauge of semiconductor stocks, along with other chip and AI infrastructure giants Nvidia, Oracle Corp., Broadcom and AMD. All of these short positions are via put options and are at least $1 billion or more in size.
Who Is Leopold Aschenbrenner
A German-born AI researcher and Columbia University valedictorian, Aschenbrenner joined OpenAI’s “Superalignment” team in 2023, tasked with ensuring superhuman AI systems remain safe and aligned with human interests.
However, his tenure ended abruptly in April 2024 when he was fired over an alleged information leak, a claim he disputes. Aschenbrenner maintains that his dismissal was actually retaliation for an internal memo warning of insufficient security and potential industrial espionage by foreign entities at OpenAI.
After OpenAI Stint
After leaving OpenAI, Aschenbrenner published a 165-page essay, “Situational Awareness: The Decade Ahead,” in which he predicted that Artificial General Intelligence (AGI) could arrive as early as 2027.
The Long Bets
While Aschenbrenner is short on big tech, he has long positions in energy-intensive infra, such as Bloom Energy ($878 million), along with Bitcoin mining firms such as IREN and Core Scientific. CoreWeave also features as one of his long positions.
As of March 31, the equity exposure of his fund has more than doubled in value to $13.6 billion from $5.5 billion at the end of December 2025.
