Leopold Aschenbrenner, the former OpenAI strategist who spent 2024 preaching the gospel of AGI, is undergoing a brutal collision with reality. His hedge fund, Situational Awareness, has traded theoretical manifests for the gritty logistics of industrial foundations. After a disastrous run in public AI infrastructure stocks that reportedly forced a massive liquidation to Ken Griffin’s Citadel, the fund is pivoting. With assets under management sliced from $20 billion to a lean $10 billion, Aschenbrenner is no longer betting on the general market tide. Instead, he is doubling down on the physical bottlenecks that actually dictate the pace of AI development: the chip supply chain.
The $400 Million Injection into Source Foundry
As reported by The Wall Street Journal, Situational Awareness has committed an additional $400 million to Source Foundry, a Stanford-born startup tasked with the unglamorous but vital job of making chip manufacturing faster and cheaper. This brings the fund's total stake in the venture to $500 million. For Aschenbrenner, this isn't just an investment; it is a desperate play for vertical integration. By targeting the primary capital expenditure pain point—the sheer cost and time-to-market of specialized silicon—Situational Awareness is attempting to buy its way into the manufacturing layer where the real power resides.
The fund invested $400 million into Source Foundry, a startup aiming to make chip manufacturing faster and cheaper, bringing its total investment in the company to $500 million.
This high-stakes gamble on Source Foundry represents a pivot from software-layer speculation to the hard reality of compute costs. While the fund has dumped its broader public holdings, it notably clung to its shares in Anthropic, signaling a retreat into a concentrated portfolio of private, high-conviction assets. The move places Aschenbrenner in a precarious position: he is now attempting to compete with established foundry giants while operating under immense financial pressure. Whether a Stanford startup can realistically disrupt the entrenched economics of semiconductor fabrication remains an open question, but for a fund that has already lost half its weight, controlling the hardware pipeline is the only remaining path to relevance. This is no longer about predicting the future; it is about owning the tools that build it.