JPMorgan Cuts Off Situational Awareness: Wall Street's Verdict on the AI Trade's Biggest Blowup

The cleanup from the AI trade’s most spectacular blowup just claimed its first major Wall Street casualty — not the fund itself, but its banker. On September 11, the Financial Times reported that JPMorgan Chase has ended its lending relationship with Situational Awareness, the AI-focused hedge fund founded by former OpenAI researcher Leopold Aschenbrenner, after large losses tied to the fund’s leveraged AI bets. Reuters confirmed the move, citing a source familiar with the matter.

The details are stark. JPMorgan, one of the fund’s key prime brokers, notified Situational Awareness that it would end its lending relationship after the losses. The move effectively revokes the credit line that once financed billions of dollars of leveraged positions in AI chip and infrastructure stocks — the same leverage that turned a spectacular winner into a case study in risk management.

A refresher on the blowup

Situational Awareness was arguably the single most concentrated expression of the AI-supremacist trade on Wall Street. Founded in 2024 by Aschenbrenner — a former OpenAI researcher and author of the “Situational Awareness” manifesto that gave the fund its name — the fund reportedly grew from roughly $225 million at launch to as much as $45 billion at its peak, including leveraged positions, after returning more than 1,000% from launch through June 2026. A July 24 investor letter reported a 439% net return through June 30, according to the Financial Times.

Then the AI trade reversed in July. The fund had built concentrated stakes in memory and storage names — including a $5.7 billion position in SanDisk and a $5.6 billion position in Micron, together more than half of its US public portfolio. When those stocks dropped sharply — SanDisk fell nearly 47% and Micron roughly 29% that month — margin calls from lenders forced a fire sale of the fund’s entire public equity book to Ken Griffin’s Citadel at a roughly 10% discount. The fund posted a 67% loss for July alone, and its asset base collapsed from roughly $45 billion to around $10 billion.

It was one of the fastest destructions of hedge fund capital in recent memory — and it put prime brokers, the banks that finance hedge funds’ leveraged trades, squarely in the regulatory spotlight.

The JPMorgan exit, decoded

JPMorgan’s exit is more than a business decision — it is Wall Street’s formal verdict on financing concentrated AI leverage. The bank was one of the fund’s key lenders, and its withdrawal removes a significant pillar of the fund’s financing structure. Reuters reports that the decision followed the losses; the FT notes the fund’s performance has lagged badly since: Situational Awareness’s funds have fallen about 4% this year, even as the S&P 500 has rallied.

There is also a regulatory shadow. In late August, the SEC subpoenaed Bank of America, Citigroup, Goldman Sachs and JPMorgan as part of an early-stage investigation into the fund’s near-collapse, seeking details on trade timing and communications with lenders about borrowed money. No one has been accused of wrongdoing, and early-stage inquiries often end without charges. But for a prime broker weighing whether to keep financing a client whose unwind is under active regulatory reconstruction, the risk calculus is straightforward — and JPMorgan appears to have made it.

The rebound nobody expected

Here is the twist: the fund is not dying. It is rebuilding. In a separate FT exclusive published September 11, the paper reported that Aschenbrenner’s firm has started building relationships with new brokerages as it attempts to rebound from what the FT calls the largest hedge fund loss in history. CNBC reported the fund has returned to buying options tied to AI stocks — a way to keep exposure to the thesis while strictly capping downside to the premium paid — and has grown back to more than $20 billion in assets under management.

That last number deserves scrutiny. A bounce from $10 billion to $20 billion-plus in six weeks implies either substantial inflows (the FT reported in late July that the fund was seeking to raise capital and had held talks with existing investors), a recovery in the value of its retained private holdings, or both. The fund retained its private stake in Anthropic — reportedly targeting an IPO at a valuation of up to $2 trillion — which remains its most valuable asset. Aschenbrenner’s wife serves as chief of staff to Anthropic CEO Dario Amodei, a connection that has drawn attention throughout the saga.

The pivot to options is the tell. After a blowup driven by borrowed money, a fund that returns to the same theme via defined-risk instruments is signaling: the thesis hasn’t changed, but the risk management has. Options on AI stocks let Situational Awareness re-express conviction without margin calls — the exact mechanism that destroyed it in July.

Why this matters beyond one fund

Three threads worth watching.

Prime brokerage is the AI trade’s soft underbelly. The Situational Awareness saga has shown that the financing layer — the banks extending leverage against concentrated AI books — is where stress propagates fastest. JPMorgan’s exit sets a precedent: when an AI-focused fund blows up, its financing doesn’t quietly get restructured; it gets withdrawn. Expect other banks to reprice or exit similar relationships preemptively.

The SEC investigation now has a moving target. Regulators are reconstructing the collapse trade by trade while the fund actively re-grows, courts new brokers, and re-enters the market through derivatives. How the fund discloses its new positioning to new counterparties will be exactly the kind of question investigators care about.

The market is still funding AI conviction — just with guardrails. A fund that lost 67% in a month has re-raised to $20 billion-plus within weeks. That says something bracing about the depth of conviction in the AI trade — and about how quickly capital forgives when the underlying theme is deemed structural. The lesson of 2026’s loudest blowup may ultimately be that leverage, not AI, was the fatal variable all along.

For a fund named after a manifesto predicting superintelligence by decade’s end, the irony cuts both ways: the thesis keeps attracting capital, but Wall Street’s patience with financing it on margin has officially run out. JPMorgan walked; someone else will presumably lend. The question the market now gets to answer in real time is whether the next financing cycle of the AI trade comes with better brakes.