Signals Inbox·August 25, 2026·FinTech

Why is the SEC probing Situational Awareness?

The SEC is probing Situational Awareness because a spectacularly leveraged AI trade turned into a multi-bank liquidity crisis, forcing a rapid portfolio sale to Citadel and giving regulators plenty to reconstruct around leverage, disclosures and trade timing.

Q1What exactly is the SEC probing at Situational Awareness?

The SEC is currently trying to reconstruct how Situational Awareness used borrowed money, traded through its July crisis and communicated with the Wall Street banks financing those trades.

The latest reporting from the New York Times and the Financial Times says the SEC has subpoenaed Bank of America, Citigroup, Goldman Sachs and JPMorgan. Regulators asked for information on the timing of Situational Awareness's trades and its communications with lenders about the money it was borrowing. The banks were also told to preserve records related to the fund.

The scope gives us a fairly good idea of what regulators care about. Situational Awareness had become one of Wall Street's fastest-growing hedge funds, made highly leveraged bets around the AI boom, suffered a brutal reversal and then sold most of its public portfolio to Citadel under pressure from lenders.

As of now, the SEC has made no public accusation of fraud, insider trading or market manipulation. The inquiry is still at the information-gathering stage. Regulators are trying to establish exactly how the fund reached the point where lenders were demanding cash and a rival hedge fund had to absorb most of its liquid portfolio.

Q2Why did one terrible month become an SEC problem?

Situational Awareness attracted the SEC because its July loss became a financing event involving several major banks and billions of dollars of forced trading.

The fund lost about 67% during July, according to an investor letter reviewed by the Financial Times. A large loss alone would normally be a problem for the fund and its investors. Here, leverage changed the situation. Goldman Sachs, JPMorgan, Bank of America and other lenders were financing positions large enough that falling asset prices led to margin pressure and urgent demands for liquidity.

Situational Awareness eventually sold the bulk of its public-equity portfolio to Citadel at roughly a 10% discount, according to the Financial Times. The transaction came together in about a day.

At that point, a private investment mistake becomes interesting to regulators. The fund's ability to hold its positions had become dependent on decisions made by prime brokers. Those decisions then affected when billions of dollars of stocks could be sold and who would have to absorb them.

The SEC already monitors large hedge funds precisely for events such as major investment losses, margin stress, counterparty problems and disruptions in prime-broker relationships. Situational Awareness managed to combine several of those problems in one episode.

Q3How big did Situational Awareness actually get?

Situational Awareness became a tens-of-billions-of-dollars hedge fund in roughly two years, although the exact peak depends on which measure and point in time we use.

The New York Times says Situational Awareness managed more than $30 billion at its peak and borrowed tens of billions more. The Wall Street Journal reported a higher figure of roughly $45 billion around the beginning of July. We would avoid pretending those numbers are perfectly comparable because fund NAV, managed assets and gross investment exposure can move quickly in a highly leveraged portfolio.

The public filings still show how extraordinary the growth was. Situational Awareness's first 13F, covering the end of 2024, showed about $255 million of reportable U.S. positions. Its latest filing, covering the end of June 2026, showed $20.24 billion.

That is roughly a 79-fold increase in reported 13F value in six quarters.

The 13F is only part of the portfolio. It excludes many private investments, foreign securities and other exposures, while options can complicate comparisons. We therefore use it to measure the speed of the expansion rather than total fund size.

Even on that narrower measure, Situational Awareness went from a small new manager to controlling public positions measured in tens of billions remarkably quickly.

Situational Awareness disclosed U.S. portfolio growth

Q4How concentrated was Situational Awareness before the crash?

By the end of June, Situational Awareness had put 55.6% of its disclosed U.S. portfolio into just SanDisk and Micron.

The latest 13F, filed after the turmoil, gives us the clearest snapshot yet of how concentrated the fund had become immediately beforehand. Situational Awareness reported $5.67 billion in SanDisk and $5.57 billion in Micron. Bloom Energy added another $1.90 billion, followed by $1.27 billion in TSMC and $1.23 billion in Nebius.

We calculate that those five positions represented about 77% of the entire disclosed portfolio. The ten largest were close to 92%.

More revealingly, the fund had become much more exposed to outright equity moves. Its March filing contained about $8.46 billion of reported put-option positions, including bearish exposure involving Nvidia, Oracle, Broadcom and other technology names. By the end of June, almost all of those puts had disappeared from the 13F; the filing showed only a tiny remaining put position relative to the size of the portfolio.

So the fund entered July with more capital in direct AI-infrastructure stocks and far less visible option protection than it had three months earlier.

SanDisk subsequently fell roughly 47% during July and Micron about 29%. Those two stocks were already more than half of the disclosed book before leverage entered the calculation.

Largest disclosed positions before the July crash

Q5How much leverage was Situational Awareness using?

Situational Awareness was using enough leverage that a sharp market correction could quickly become a cash crisis.

The New York Times reports that the fund borrowed tens of billions of dollars on top of more than $30 billion under management. CNBC reporting has put leverage as high as roughly 400% at points before the unwind.

We should treat the 400% figure carefully because leverage can be calculated in several ways. Gross exposure, borrowed cash, derivatives and exposure relative to NAV can produce very different percentages. The more reliable evidence is what actually happened: several large banks were financing the fund, those lenders demanded additional collateral as positions fell, and Situational Awareness had to raise cash fast.

The strategy had worked spectacularly while prices moved in the fund's favor. Situational Awareness returned about 439% after fees through the first half of 2026, according to the Financial Times. Leverage magnified those gains and increased the amount of capital the fund could deploy into the same AI thesis.

July showed the other side of the trade. Once collateral values fell, lenders cared about today's balance sheet rather than where AI infrastructure stocks might trade several years later. Situational Awareness's investment horizon suddenly became much shorter than its investment thesis.

Q6What actually broke inside Situational Awareness?

Situational Awareness's liquidity broke when concentrated AI positions, bearish technology trades and heavy borrowing started hurting the fund at the same time.

The public portfolio had become heavily exposed to memory, chips, data centers, compute and power. SanDisk, Micron, Bloom Energy, Nebius, CoreWeave, Applied Digital and Core Scientific all expressed different versions of the same broad AI-infrastructure boom.

During July, many of those trades reversed sharply. Meanwhile, reporting on the fund's strategy indicates that some bearish positions elsewhere in technology also moved against it. The portfolio therefore lost money on exposures that were supposed to behave differently.

Borrowing made each move harder to absorb. Falling securities reduced the collateral supporting the fund's financing. Banks asked for more collateral. Raising collateral required selling positions. Selling into weak markets put more pressure on prices and on the fund's remaining equity cushion.

There was another complication: some of Situational Awareness's most valuable assets were private. The fund had built a major Anthropic position, but a private AI stake cannot be turned into cash as easily as publicly traded Micron shares when a prime broker wants collateral immediately.

Situational Awareness eventually kept much of its private portfolio while unloading most of the liquid public book. It also tells us what actually failed. The fund still owned valuable assets, but it needed cash faster than those assets could provide it.

Q7Did Situational Awareness hit the SEC's hedge-fund stress triggers?

Situational Awareness appears to have experienced exactly the kind of extreme-loss and margin-stress events covered by the SEC's current Form PF reporting system.

The relevant rules come from the SEC's 2023 Form PF amendments, which have already taken effect for large hedge-fund event reporting. They require qualifying advisers to report certain major events as soon as practicable and within 72 hours.

One trigger covers an extraordinary investment loss of at least 20% of a qualifying fund's reporting-fund aggregate calculated value over a rolling ten-business-day period. Other triggers cover significant margin events, counterparty defaults, major prime-broker restrictions and certain redemption problems.

We cannot see Situational Awareness's confidential Form PF filings, so we cannot determine which technical triggers it reported or exactly when. The public numbers still place the July episode far outside normal volatility. As seen above, the fund lost roughly two-thirds during the month and was simultaneously dealing with margin pressure from major lenders.

That makes Form PF directly relevant to the SEC's reconstruction of what happened. Regulators may already possess confidential reports from the fund describing stress events that outsiders only learned about later through the press.

Q8Why is the SEC subpoenaing Goldman, JPMorgan, Citi and Bank of America?

The banks can tell the SEC what Situational Awareness was saying about its leverage and liquidity while regulators compare those statements with what was actually happening across the whole fund.

Situational Awareness had financing relationships with several major institutions. Each bank would have records covering its own loans, collateral requirements, margin calls, trading activity and conversations with the fund.

That gives the SEC something concrete to test. If Situational Awareness told Goldman one thing about available liquidity, JPMorgan another thing about borrowing elsewhere, and regulators something different on Form PF, those records could expose inconsistencies. If everything matches, the same records could show that the fund was simply taking enormous risks with lenders who understood what they were financing.

The multi-bank structure itself deserves attention. A prime broker generally sees its direct relationship with a client far more clearly than it sees every position and financing arrangement held at competing banks. A hedge fund can therefore build a much larger aggregate exposure than any individual lender is financing.

This dynamic became notorious after Archegos. We currently have no evidence that Situational Awareness concealed positions from its banks as Archegos was later accused of doing. The SEC's subpoenas are one way to find out whether everyone involved really had the same picture.

The fund's investor communications may also be reviewed. The Wall Street Journal reported that some investors had pushed for more transparency as Situational Awareness grew. Any statement about leverage, liquidity or risk can now be compared with the fund's internal records and its communications with lenders.

Q9Why does the SEC want the exact timing of Situational Awareness's trades?

The timing can show whether Situational Awareness was voluntarily cutting risk or selling because its lenders had effectively taken control of the situation.

The SEC can line up several clocks: when AI stocks started falling, when individual banks raised margin requirements, when the fund sought fresh capital, when positions were sold, when Citadel entered negotiations and what Situational Awareness told counterparties along the way.

The timeline can answer a very practical question. Did the fund still have meaningful freedom to choose when and what to sell, or had margin pressure already forced its hand?

Trade timing also helps regulators measure market impact. A fund liquidating billions of dollars of concentrated positions can push prices further against itself, particularly when other traders know which names it owns. Aschenbrenner told investors that market participants were trading against publicly associated Situational Awareness positions and compared the dynamic to a bank run.

Bloomberg later reported that short-selling data did not support a simple story in which predatory shorts alone brought the fund down. That makes the chronology even more useful. Regulators can separate the effect of ordinary market selling, the fund's own forced trades and activity by other investors rather than relying on one explanation from the middle of the crisis.

Q10Is the SEC accusing Situational Awareness of fraud or insider trading?

As of now, the public evidence points to a broad investigation of leverage, trading and disclosures rather than an announced fraud or insider-trading case.

That is worth stating clearly because Situational Awareness comes with all the ingredients that make people jump to more dramatic conclusions. Leopold Aschenbrenner previously worked at OpenAI, the fund invested heavily across the AI ecosystem, it owned a large Anthropic stake, and Aschenbrenner has close personal connections inside the industry.

None of those facts establishes illegal trading.

The latest reports say regulators are collecting information about trades and communications with lenders. Bloomberg has also stressed that an SEC inquiry can remain an information-gathering exercise and can end without an enforcement action.

The situation would change quickly if regulators found false statements to lenders, misleading regulatory filings, hidden exposures or trading based on material non-public information. There is currently no public finding along those lines.

The line is pretty simple: the SEC has a serious reason to investigate how Situational Awareness financed and unwound its portfolio, but the evidence available today does not support calling this a fraud case.

Q11Did Situational Awareness's collapse hurt anyone beyond its own investors?

Yes. Situational Awareness became large enough that its July problems spilled into other major Wall Street firms and into the trading of AI-related stocks.

Jane Street gives us the clearest example. Reuters reported that the trading firm took roughly a $15 billion hit during July from its exposure to Situational Awareness and from other technology positions hurt by the same selloff. Jane Street told employees that its investment in Situational Awareness had suffered a large drawdown, while separate market positions also contributed.

We should therefore avoid saying Situational Awareness alone cost Jane Street $15 billion. The useful observation is broader: one of the world's most sophisticated trading firms had enough overlapping exposure to the fund and the same market theme that July became its first negative trading-revenue month in roughly a decade.

The threat of further forced selling also became part of the market itself. Situational Awareness held multi-billion-dollar positions in names already falling quickly. Had those positions simply been dumped into the open market, the fund could have added another large price-insensitive seller to an already violent AI correction.

Citadel's intervention reduced that immediate pressure by moving much of the portfolio into a stronger balance sheet.

That is why the SEC's interest extends beyond whether Situational Awareness investors knowingly accepted a risky strategy. The episode reached counterparties and public markets too.

Q12Why was the Citadel rescue such a big deal?

Citadel's purchase gave Situational Awareness enough room to repay lenders without trying to dump the entire public portfolio into a falling market.

The deal happened fast. According to Bloomberg and the Financial Times, Citadel negotiated the transaction in roughly 24 hours and bought the portfolio at around a 10% discount. Millennium and other firms were also interested.

The freshest evidence shows what Citadel actually did with that risk afterward. Ken Griffin told Citadel investors that the firm had removed more than 80% of the aggregate risk it acquired from Situational Awareness within about three weeks. Citadel executed nearly 100 block trades worth more than $4 billion, including unusually large trades in several individual stocks.

The original portfolio was too large and concentrated for even Citadel to simply leave untouched. Citadel could warehouse the positions, wait for better liquidity and distribute them progressively instead of forcing Situational Awareness to sell everything at once.

Several former Situational Awareness holdings rebounded after the deal, and Citadel's flagship Wellington fund gained about 5.9% in July, its best month in years. We cannot attribute that entire gain to one transaction, but the economics of buying distressed assets at a discount and selling them into a recovering market were clearly favorable.

For regulators, the rescue also leaves a detailed trail of prices and transactions that can show how much stress had built up before the transfer.

Q13Is Situational Awareness basically another Archegos?

Situational Awareness has enough in common with Archegos to justify the comparison, but the known evidence stops well short of making the two cases equivalent.

Archegos also built huge concentrated positions using financing from several Wall Street banks. When those positions fell, margin calls led to forced liquidation and billions of dollars of losses at counterparties. Credit Suisse alone eventually lost about $5.5 billion.

The SEC later alleged that Archegos founder Bill Hwang and colleagues had lied to counterparties about the firm's exposure, concentration and liquidity and had manipulated stocks. That alleged deception became central to the enforcement case.

Situational Awareness currently sits at an earlier and very different point. The fund is an SEC-registered investment adviser. Regulators already receive information through Form ADV, Form PF and other filings. Public reporting has established concentration, heavy borrowing, margin pressure and forced selling. It has yet to establish deception.

Archegos is useful mainly as a map of what regulators will want to check. The SEC knows from experience that several banks can each believe they understand their own exposure while missing the scale of the client's combined risk.

Situational Awareness compared with Archegos

Q14Is the SEC really investigating Situational Awareness because it bet on AI?

The SEC's current probe is about the way Situational Awareness financed and unwound its AI bets, while the underlying AI thesis mainly explains why the portfolio became so concentrated.

Aschenbrenner built the fund around a clear view: increasingly powerful AI would require enormous amounts of compute, memory, data-center capacity and electricity. Situational Awareness expressed that view through companies including SanDisk, Micron, Bloom Energy, CoreWeave, Nebius and TSMC, as well as private investments such as Anthropic.

That thesis worked extraordinarily well for a long time. The fund reportedly gained more than 1,000% from inception through the end of June.

The SEC has no reason to decide whether Aschenbrenner's prediction about AI infrastructure is intellectually correct. Regulators care that the fund combined the thesis with huge position sizes and borrowed money from several systemically important financial institutions.

Anthropic also shows why investment value and usable liquidity can diverge. Situational Awareness retained its private investments through the crisis while selling public stocks. A multi-billion-dollar private position can make a fund wealthy on paper and still be a poor source of cash during a fast-moving margin call.

The regulatory story comes from the financing structure around the AI thesis, not from the thesis itself.

Q15Could the banks financing Situational Awareness have their own problem?

Goldman Sachs, JPMorgan, Citi and Bank of America could face uncomfortable questions about risk controls even if the SEC never accuses any bank of wrongdoing.

Prime brokerage has become a very lucrative Wall Street business. Financial News recently calculated that major institutions generated about $22.5 billion of prime-brokerage revenue during the first half of 2026, up roughly 51% from the previous year.

Banks are also competing aggressively for hedge-fund balances. Citi has discussed expanding its prime balances from roughly $450 billion toward $700 billion by 2028, while Goldman, JPMorgan and Morgan Stanley have each operated with more than $1 trillion.

Situational Awareness was precisely the kind of client bankers want: fast-growing, highly active, trading complicated products and requiring large amounts of financing.

The SEC can now compare how several banks handled that client as its exposure expanded at extraordinary speed. Regulators can see what limits were imposed, what collateral was requested, how the banks estimated concentration risk and whether lenders understood how much financing Situational Awareness was receiving elsewhere.

There may be a perfectly ordinary explanation: banks financed a sophisticated hedge fund, marked collateral as markets moved and protected themselves by making margin calls. The subpoenas give regulators enough information to see whether that explanation holds.

Q16Why is the Situational Awareness probe awkward for the SEC's Form PF overhaul?

Situational Awareness has become a timely example of the systemic-risk scenario behind Form PF just as the SEC is trying to make parts of that reporting regime lighter.

The SEC and CFTC proposed a major Form PF simplification in April 2026. Among other changes, they want to raise the threshold for a "large hedge fund adviser" from $1.5 billion to $10 billion and simplify or remove several reporting requirements.

Situational Awareness would easily clear even the proposed $10 billion threshold, so this specific fund would remain in the group receiving the most scrutiny. Still, the broader timing is striking.

When the SEC introduced its 2023 event-reporting rules, it specifically highlighted extraordinary investment losses, margin problems and counterparty defaults as events regulators needed to see quickly because they could point to systemic stress. Situational Awareness has now delivered a real-world example involving huge losses, several prime brokers and a forced transfer of a multi-billion-dollar portfolio.

The later 2024 overhaul of other Form PF reporting requirements has a separate timeline and is currently delayed until October 2026. The new 2026 proposal would simplify the regime further if adopted.

We cannot infer that Situational Awareness will change the SEC's policy direction. It does, however, give both sides of the Form PF debate a fresh case to argue over: how much information does the government really need before a leveraged hedge-fund problem becomes visible through market prices and emergency sales?

Q17What happens next in the SEC investigation of Situational Awareness?

The SEC will probably spend the next phase comparing records rather than rushing toward charges.

Regulators now have several versions of the same event available to them. Situational Awareness has its internal books and communications. Each lender has margin records, financing agreements and messages with the fund. Citadel has transaction records from the portfolio transfer. The SEC also has confidential regulatory filings that the public cannot see.

The investigation becomes much more serious if those sources fail to match.

A discrepancy over total leverage, available liquidity, collateral, exposure at other banks or the timing of forced trades could give regulators a specific conduct issue to pursue. Matching records would support a much simpler conclusion: Situational Awareness ran an extraordinarily aggressive strategy, markets moved violently against it, lenders protected themselves and the fund survived by deleveraging.

The fund is still operating today. Aschenbrenner told investors that Situational Awareness had removed leverage, and recent reporting says it continues to own valuable private positions. It has even returned to making private investments after the July crisis.

The next public development should be easy to read. Another subpoena would tell us relatively little. A specific alleged false statement, reporting failure or securities-law violation would tell us that the SEC has moved from reconstructing the collapse to building a case.

Q18So why is the SEC probing Situational Awareness?

The SEC is probing Situational Awareness because extreme concentration and heavy borrowing turned one hedge fund's bad trades into a multi-bank liquidity crisis with consequences across Wall Street.

We can now see the sequence fairly clearly. Situational Awareness grew from roughly $255 million of disclosed U.S. positions at the end of 2024 to more than $20 billion by the end of June 2026. Its two biggest reported holdings then represented 55.6% of that public portfolio. At the same time, the fund had largely removed the put positions visible three months earlier and was borrowing tens of billions of dollars across major Wall Street firms.

When the AI trade reversed, that structure left very little room for error. Falling positions triggered margin pressure, liquidity became the immediate problem, and Situational Awareness sold most of its public book to Citadel at a discount. Citadel subsequently needed nearly 100 block trades worth more than $4 billion to shed most of the risk it had taken on. Jane Street, meanwhile, suffered part of a roughly $15 billion monthly loss through its own exposure to Situational Awareness and the same market rout.

The SEC now wants to know what happened between those points: what the fund told its banks, how much each lender understood about the overall leverage, when the fund began selling and whether regulatory filings accurately captured the stress.

The conclusion is pretty specific. The Situational Awareness probe currently looks like an investigation into whether a huge leveraged hedge-fund unwind was handled and disclosed properly. The collapse itself gives the SEC more than enough reason to investigate. Evidence of an actual securities-law violation has yet to emerge publicly.