Six weeks is not a long time to rebuild conviction after watching tens of billions of dollars in positions evaporate. Yet Leopold Aschenbrenner, the former OpenAI researcher who founded the hedge fund Situational Awareness, has returned to the exact trade that nearly destroyed him — loading up on call options across some of the most volatile names in the artificial intelligence supply chain.
According to a CNBC report citing sources familiar with the matter, Situational Awareness executed new call-option positions on Advanced Micro Devices (AMD), Bloom Energy, CoreWeave, SK Hynix, and SanDisk. The trades were placed across a window stretching from late last week through this week — a deliberate, phased re-entry rather than a single impulsive wager. The move comes roughly six weeks after an AI-linked market dislocation that produced what sources described as a $35 billion wipeout for the fund's positions.
The anatomy of a conviction trade
What makes Aschenbrenner's re-entry significant is the composition of the basket. These are not diversified, hedged bets spread across sectors. They are targeted, leveraged positions — call options — on companies sitting directly inside the artificial intelligence infrastructure stack. AMD supplies the graphics processing units that power model training at scale. CoreWeave operates the GPU cloud infrastructure that hyperscalers and AI labs increasingly depend on for on-demand compute. SK Hynix is the dominant supplier of high-bandwidth memory, the scarce silicon that allows modern AI accelerators to function at peak throughput. Bloom Energy provides the power generation capacity that data centers increasingly need as AI workloads push electricity consumption to record levels. SanDisk, historically a flash-storage brand, rounds out the exposure to the hardware layer of the AI buildout.
Together, the portfolio reads less like a diversified fund and more like a single, high-conviction thesis: that the infrastructure buildout underpinning the AI wave is not finished, and that the $35 billion drawdown six weeks ago represented a buying opportunity rather than a structural breakdown. Call options amplify that bet — they offer leveraged upside if the underlying stocks rally, while theoretically capping losses at the premium paid if the thesis fails again.
Who is Leopold Aschenbrenner?
Aschenbrenner is not a typical hedge fund manager. He spent time at OpenAI, one of the most closely watched artificial intelligence laboratories in the world, before leaving to launch Situational Awareness — a fund whose very name signals that its strategy is built around anticipating macro-level technological shifts rather than trading quarterly earnings cycles. His background gives him a degree of technical credibility unusual in the hedge fund world, and his public writing on AI timelines and geopolitical risk has attracted significant attention in both Silicon Valley and Washington policy circles.
That credibility, however, did not protect the fund from the brutal AI market sell-off that produced the $35 billion position wipeout. Markets have a way of punishing concentrated conviction regardless of how well-reasoned the underlying thesis appears. The question now is whether the near-destruction of the fund six weeks ago has sharpened Aschenbrenner's timing — or simply confirmed a temperament that prioritizes long-term structural bets over near-term risk management.
Infrastructure as the enduring bet
For readers in the digital-asset space, the parallels to crypto infrastructure cycles are hard to ignore. The pattern — a violent drawdown in speculative infrastructure names followed by a rapid re-accumulation by high-conviction players — has repeated across blockchain infrastructure, Layer-2 networks, and mining hardware sectors. The specific assets differ, but the psychological and financial mechanics are nearly identical. Early, concentrated bettors get hurt badly in the correction; those who survive with capital intact often re-enter at exactly the point when broader sentiment is most negative.
Aschenbrenner's re-entry also signals something about where institutional money sees the AI cycle. If the thesis were dead — if the $35 billion wipeout had reflected a fundamental mispricing of AI's commercial trajectory — a rational actor would rotate out entirely. Instead, Situational Awareness is buying the same infrastructure layer again, suggesting the fund views the earlier dislocation as a liquidity event rather than a verdict on AI's commercial viability.
What this means
The immediate market takeaway is that at least one technically credentialed, institutionally significant player views the AI infrastructure correction of six weeks ago as an entry point rather than an exit signal. Whether that judgment proves correct will depend on factors ranging from Federal Reserve policy to the pace of hyperscaler capital expenditure to geopolitical constraints on semiconductor supply chains involving SK Hynix and other Korean chipmakers. What is clear is that Aschenbrenner has made his read public through his trades: the buildout continues, the infrastructure names are undervalued relative to the long-term demand curve, and the $35 billion wipeout was a price worth paying for another seat at the table. Conviction of that magnitude — or recklessness of that scale, depending on the outcome — rarely goes unnoticed by the broader market.
Written by the editorial team — independent journalism powered by Bitcoin News.