Did One Hedge Fund Cause the July AI Rout? Fact-Checking Situational Awareness and the SK hynix ADR

A compelling explanation spread after AI infrastructure stocks staged an extraordinary rebound at the end of July 2026. The story was that Situational Awareness, the hedge fund led by former OpenAI researcher Leopold Aschenbrenner, had taken a $7 billion cornerstone position in the SK hynix ADR, received margin calls, and transferred a $16 billion public-equity book to Citadel. The fund’s liquidation supposedly explained both the July selloff and the July 31 rebound.

That narrative mixes confirmed facts, credible reporting, a material misreading of the SEC prospectus, and causal claims that cannot yet be proved. The best-supported conclusion is narrower: a large transfer of Situational Awareness’s public-equity book appears to have occurred, and forced deleveraging likely amplified moves in crowded AI infrastructure stocks. But the fund did not publicly commit to a $7 billion SK hynix allocation by itself, and no public dataset can attribute 40% to 50% of the KOSPI decline to one fund.

Data cutoff: U.S. returns use the July 30, 2026 close. Korean prices are intraday snapshots at 12:12 p.m. KST on July 31. Private-fund AUM, leverage, prime-broker relationships, and margin calls have limited public disclosure, so this report separates SEC filings from media reports and inference. It follows our July 28 Korean memory selloff analysis and eight questions behind the AI CAPEX correction. More research is available in the AI and HBM hub and Exclusive Analysis hub.

TL;DR

• The Wall Street Journal and Axios reported that Situational Awareness sold most or all of its public-equity portfolio to Citadel after large losses. The direction of the transaction is supported by credible reporting, although the exact value and terms are not publicly filed.
• The SEC prospectus does not say Situational Awareness alone received a $7 billion SK hynix allocation. Baillie Gifford, Coatue-managed funds, and Situational Awareness collectively expressed non-binding interest in up to $7 billion. Individual allocations were not disclosed.
• Q1 2026 filings confirm meaningful exposure to Bloom Energy, CoreWeave, and SanDisk. A May Schedule 13G confirms a 5.6% Nebius stake. Micron and Nvidia exposure, however, included substantial options. The portfolio was not a simple long-only AI basket.
• On July 30, NBIS rose 27.13%, BE 26.49%, SNDK 25.99%, CRWV 21.51%, MU 18.36%, and SKHY 17.52%, versus 2.65% for NVDA. This is consistent with forced-seller removal and short covering, but it is not a controlled experiment proving a single-fund cause.
• A separate claim that Korea suffered KRW 37.7 trillion of forced margin liquidation is also incorrect. That figure appears to refer to outstanding margin-credit balances, not executed forced sales.
• The most defensible verdict is that Situational Awareness was likely an important marginal seller and amplifier inside a leveraged, crowded factor. Its book transfer can support short-term normalization, but it does not answer the fundamental question of AI and memory earnings after 2028.

1. Claim-by-claim verdict

2. The central correction: $7 billion was not one fund’s allocation

SK hynix’s final SEC prospectus covered 177.9 million ADSs at $149 each, for a gross offering of roughly $26.5 billion and estimated net proceeds of about $26.2 billion.

The prospectus named three investors:

• Baillie Gifford Overseas Limited
• funds managed by Coatue
• Situational Awareness Partners LP

They had, severally and not jointly, indicated interest in purchasing up to an aggregate of $7 billion of ADSs. The indications were explicitly non-binding, and each investor could buy more, less, or none. SK hynix final SEC prospectus

The public record therefore establishes only:

Aggregate indication ceiling from three investors: $7 billion
Individual allocations: undisclosed
Situational Awareness allocation: undisclosed
Binding commitment: no

Situational Awareness may have received a large allocation. It may also have received far less than $7 billion. Treating the aggregate indication as a single-fund position inflated every downstream conclusion about leverage, the margin call, and market impact.

3. What is actually known about the fund

3.1 The public-equity transfer

The Wall Street Journal reported that Situational Awareness sold the bulk of its stock portfolio to Citadel after deep AI-related losses. Axios cited one source saying it sold the entire public-equity portfolio.

The source discrepancy matters. The safest wording is that most or all of the public-equity book was transferred. That does not establish that the fund itself was liquidated. Private holdings and separate vehicles may remain, and no public announcement says the manager has ceased operations.

Some secondary reports put the transferred book at $16 billion. There is no public transaction filing showing whether this figure was market value, gross exposure, or some other measure. The trade’s existence is more reliable than the exact amount.

3.2 AUM, returns, leverage, and margin calls

Media reports described assets rising above $20 billion and then falling to about $10 billion. Other reports cited a 439% net return through June, more than 1,000% since inception, up to four times leverage, and margin calls from multiple prime brokers.

These are reported figures, not values that can be reconstructed from public SEC filings. Form 13F does not disclose:

• net asset value
• cash and borrowings
• gross and net leverage
• OTC derivatives and swaps
• prime-broker collateral
• monthly P&L and redemptions
• private-asset valuations

The media reporting may be accurate, but it should not be presented as audited public fact.

4. What the 13F and 13G filings show

Situational Awareness’s Q1 2026 13F contained 42 reportable entries with aggregate reported value of about $13.68 billion. Option entries can reflect the market value of underlying securities rather than option premium or delta exposure. The total is therefore neither AUM nor a measure of net risk. Situational Awareness Q1 2026 13F

A more recent Schedule 13G reported 12,410,060 Nebius shares, or 5.6%, as of May 19. Nebius Schedule 13G

These filings support the view that the fund had large AI infrastructure, power, and storage exposure. They also show why a simple held-versus-not-held table is insufficient. Micron included both calls and puts, and Nvidia was associated with substantial put exposure.

SKHY cannot be verified from the Q1 filing because the ADR listed in July. A broad Q3 holdings check will not arrive until mid-November. A Schedule 13D or 13G could appear earlier if an investor crossed the 5% threshold, but current public filings do not disclose Situational Awareness’s allocation.

5. The rebound: strong evidence, not a controlled experiment

U.S. close on July 30

Korea intraday at 12:12 p.m. KST on July 31

The cross-section is consistent with the removal of a forced seller. It may also reflect short covering and options hedging in the names that had suffered the largest declines.

It is not a controlled experiment for five reasons:

1. NBIS, BE, and CRWV have much higher beta than Nvidia.
2. A block transfer to Citadel is different from all shares being sold in the open market.
3. Short covering and dealer gamma can magnify the rebound.
4. The fund’s net exposure is unknown because strikes, maturities, deltas, and swaps are not disclosed.
5. FOMC news, hyperscaler earnings, macro risk appetite, and Korean program flows moved at the same time.

The strongest defensible conclusion is that the rebound supports a forced-flow amplification hypothesis. It does not identify the original trigger or measure one fund’s share of the decline.

6. The KRW 37.7 trillion margin-liquidation error

Outstanding margin credit, collateral shortfalls, forced-sale orders, and executed forced sales are different statistics.

The publicly cited KRW 37.7 trillion figure appears to refer to outstanding margin-credit balances, not executed forced sales over July 24 to 28. Treating the entire stock of credit as liquidation volume materially exaggerates market impact.

The same caution applies to ELS knock-in hedging. Futures selling alone cannot identify ELS hedges without product-level delta and gamma data.

7. A more defensible causal chain

The following mechanism fits the evidence:

High-beta AI infrastructure stocks decline
→ collateral capacity deteriorates in leveraged portfolios
→ public-equity exposure is reduced or block-transferred
→ crowded-factor pressure intensifies
→ removal of the urgent seller becomes public
→ short covering, dealer hedging, and high-beta rebound

What remains unknown is crucial:

• Situational Awareness’s actual SKHY allocation and purchase price
• the position that triggered the first margin call
• prime-broker liquidation timing and size
• security-level quantities and hedges transferred to Citadel
• daily July net sales by the fund
• actual Korean forced-sale and ELS hedge execution

SKHY may have contributed to the stress. But the public record does not prove that it was the trigger. Losses on multiple AI infrastructure longs, software shorts moving against the fund, and derivatives collateral could have combined.

8. Why a 40% to 50% market attribution is not possible

To assign a share of the KOSPI or AI-factor decline to one fund, an analyst would need:

• security-level sale quantities and execution prices
• SKHY ADR versus 000660 arbitrage flows
• futures, options, and swap hedges
• executed Korean margin-liquidation data
• issuer-level ELS hedge data
• high-frequency event windows
• a credible counterfactual group

None is publicly complete. Larger drawdowns and rebounds in fund-linked names show an amplification channel, not a market-wide contribution percentage.

9. Investment implications

9.1 What improved

If the public-equity transfer occurred as reported, an urgently liquidity-constrained marginal seller is likely less active. That supports a reversal of the portion of the July decline unrelated to fundamentals. The extraordinary rebound suggests that normalization had begun.

9.2 What did not improve

The following questions remain open:

• Can AI token demand absorb memory supply growth after 2028?
• Can HBM4 and HBM4E market share and pricing hold?
• Will commodity DRAM and NAND move into oversupply?
• Can hyperscaler cash flow fund CAPEX and long-term purchase commitments?
• How will Chinese memory expansion and U.S. controls split global pricing?

The rebound is a flow signal, not an earnings validation.

10. Citadel is not permanent, unlevered capital

Describing Citadel as unlevered patient capital would be inaccurate. Citadel is a sophisticated hedge fund and trading ecosystem capable of rapidly pricing and hedging risk.

The positive interpretation is narrower: positions moved from a seller that reportedly needed immediate liquidity to a much larger risk absorber. Citadel can still:

• hedge with single-stock, index, and options positions
• distribute shares as liquidity returns
• run relative-value trades
• retain some positions and sell others

The overhang may have shifted from urgent to manageable. It did not necessarily disappear.

11. What to monitor

1. Schedules 13D and 13G for any investor crossing 5% of SKHY.
2. Q3 2026 13F filings in mid-November for broad SKHY and AI infrastructure changes.
3. ADR versus local-share pricing, borrow costs, volume, and conversion economics.
4. Options open interest and implied volatility to assess whether dealer and short-covering pressure is fading.
5. Follow-on block trades that could indicate redistribution of the Citadel book.
6. 2027-2028 earnings estimates, the final test of whether price normalization is supported by cash flow.

12. Red Team

The fund was an effect, not a cause

AI infrastructure may have been repriced because of valuation, rates, power constraints, and vendor-financing concerns. Situational Awareness could have been a casualty of that move rather than its origin.

Forced selling may not be over

Citadel may hedge or distribute the acquired positions, while other leveraged funds could face margin calls. One transfer does not cleanse an entire crowded factor.

The rebound may be mostly short covering

A one-day 18% to 27% rebound can be driven by short covering and options gamma. If volume fades and prices revisit prior lows, the normalization interpretation weakens.

Invalidation

• Future SEC filings show little or no SKHY position at Situational Awareness.
• The public-equity sale reports are denied by the parties.
• Large follow-on block sales occur after the Citadel transfer.
• AI infrastructure and memory earnings estimates decline structurally.

13. Evidence classification

[Fact]

• SK hynix offered 177.9 million ADSs at $149.
• Three investors collectively indicated non-binding interest in up to $7 billion.
• Q1 filings show BE, CRWV, SNDK, MU, and NVDA exposures.
• A May Schedule 13G reported a 5.6% NBIS stake.
• AI infrastructure and Korean memory stocks rebounded sharply.

[Reported]

• Most or all of the public-equity book was transferred to Citadel.
• Fund assets fell to roughly $10 billion.
• The fund used up to four times leverage and faced margin calls.
• The transferred book was approximately $16 billion.

[Inference]

• The transfer likely reduced urgent forced-selling pressure.
• Leveraged AI infrastructure positioning likely amplified the July decline.
• Citadel may have converted an urgent overhang into a more manageable one.

[Blocked]

• Actual SKHY allocation and entry price
• Prime-broker collateral and margin-call records
• Exact transaction value, security quantities, and Citadel hedges
• Daily July net sales and total fund net exposure
• Executed Korean margin liquidations and ELS hedges
• The fund’s exact contribution to the KOSPI and AI-factor decline

Conclusion

The Situational Awareness episode is an important piece of the July AI infrastructure selloff. The reported transfer of a large public-equity book and the explosive rebound in high-beta names support the view that forced deleveraging distorted prices.

But three corrections are essential. The $7 billion was not Situational Awareness’s disclosed standalone SK hynix allocation. KRW 37.7 trillion was not verified forced-liquidation volume. And the rebound, while powerful evidence of flow normalization, was not a controlled experiment assigning 40% to 50% of the market decline to one fund.

The most accurate summary is:

Situational Awareness has not been proved to be the sole cause of the July AI rout. It was likely an important marginal seller and amplifier inside a leveraged, crowded market structure. The public-equity transfer can help short-term normalization, but the durability of AI and memory earnings after 2028 still requires separate fundamental proof.

This report is market-structure research based on public information, not a recommendation to buy or sell any security.