Situational Awareness fund invests 400 million dollars in chip startup a bet on faster and cheaper manufacturing despite the fund's losses
Listen to this article
Read by Anchor
The AI-focused hedge fund “Situational Awareness”, founded by former OpenAI researcher Leopold Aschenbrenner when he was in his mid-twenties, injected an additional $400 million into Source Foundry, a startup founded by Stanford researchers that aims to make chip manufacturing faster and cheaper. That brings the fund’s total investment in the company to $500 million, according to a Wall Street Journal report.
The investment comes a month after the fund sold most of its public portfolio to Citadel for Ken Griffin, following a drop in the fund’s assets from $20 billion to $10 billion amid a decline in AI-infrastructure stocks. The fund retained its stake in Anthropic. Aschenbrenner, who had no trading experience when he launched the fund in 2024, posted strong early returns before being hit by a correction wave.
Source Foundry focuses on the manufacturing layer: how to turn designs into actual silicon more quickly and at lower cost. This complements Aschenbrenner’s bet on the full AI stack, from models (Anthropic) to chips (Source Foundry). The message: even amid broader losses, smart capital continues to flow toward the real bottleneck, computing manufacturing.
What this means for the region:Saudi Arabia (through the Public Investment Fund and Alat Advanced Manufacturing), the United Arab Emirates (through Mubadala and the Chips Council), and Qatar are all seeking a foothold in the semiconductor value chain. An AI-focused fund’s investment in a fledgling chip-manufacturing company, despite its losses in public equities, signals that the manufacturing layer is the next strategic bet. The region that possesses capital, energy and sovereign ambition should note: smart money is not betting only on models, but on those who build the computing that runs them.
Source Foundry has not disclosed specific technical details yet, but the Stanford founders are focusing on advanced manufacturing techniques that shorten production cycles and lower cost-per-watt, the critical metric for large-scale AI data centres. If it succeeds, the model could change the economics of building AI factories that countries are racing to construct.
Takeaway:A fund founded by a researcher in safety and scaling now bets on manufacturing physics. The apparent paradox, model safety versus the speed of chip production that runs them, is in fact an integrated vision: whoever controls the computing layer controls the pace, cost and capability of all AI.