Key Takeaways
- • Jane Street sustained a $15 billion loss in July 2026, its first negative trading month since 2016, driven by an AI-stock reversal that hit both the firm's proprietary trading book and its LP stake in Situational Awareness hedge fund.
- • Year-to-date trading revenue remains above $40 billion, exceeding the $39.6 billion generated across all of 2025, demonstrating the scale of July's drawdown against sustained performance.
- • Situational Awareness, the AI-specialist hedge fund managed by former OpenAI researcher Leopold Aschenbrenner, sold its public-equity portfolio in a distressed transaction with Citadel following margin calls; the fund's stake with Jane Street ended July approximately flat on the year despite first-half gains.
- • Jane Street's hedging strategy—short-term put options designed to protect against sharp shocks—failed because July's losses were "relatively spread out throughout the month," rendering crash-insurance puts ineffective; memory and semiconductor names declined approximately 50%.
- • The incident exposes structural crowding across multiple asset classes and investment vehicles: the same AI factor trade appeared simultaneously in Jane Street's market-making inventory, its LP position in a specialist hedge fund, and pod-shop factor books, creating correlated downside when the trade reversed.
Summary
Jane Street, the $15 billion-capitalized proprietary trading firm founded in 2000, suffered its worst month in a decade during July 2026. The $15 billion loss marked the first negative trading month since 2016 and represented a roughly 25% decline from the end-June revenue peak. Despite this significant drawdown, the firm's year-to-date trading revenue stood above $40 billion—a figure that already exceeds the $39.6 billion the firm generated across the entirety of 2025, underscoring the magnitude of first-half outperformance before the July reversal.
The losses stemmed from an AI-stock correction that cascaded through multiple layers of Jane Street's portfolio simultaneously. The firm held an LP stake in Situational Awareness, the hedge fund managed by Leopold Aschenbrenner, former OpenAI researcher, which had been a strong performer in the first half of 2026. According to internal notes reviewed by Reuters, the Situational Awareness position "became large by performing well in the first half of the year," then suffered a severe drawdown that left the stake approximately flat on the year, though still profitable over the full holding period. Facing margin pressures, Situational Awareness liquidated the bulk of its public-equity portfolio through a distressed transaction with Citadel Securities.
Jane Street's own technology and semiconductor exposure compounded the impact. The firm's largest memory and semiconductor positions declined roughly 50% during July, while non-AI Asia longs that had generated strong second-quarter returns reversed sharply. The structural failure of the firm's risk management framework became apparent when its standard hedge—short-term put options purchased against tail events—proved ineffective. Because July's AI losses were "relatively spread out throughout the month" rather than concentrated in a single sharp shock, these short-term puts provided minimal protection. The firm noted it "largely lost on the same portfolio of trades that had strong outperformance in the second quarter."
The deeper concern for allocators involves the concentration of the AI trade across seemingly diversified asset classes. Jane Street simultaneously operated as a market-maker providing liquidity in AI-related securities, an LP in an AI-specialist hedge fund with overlapping factor exposure, and a proprietary trader holding correlated long positions. When the reversal occurred, these positions amplified rather than hedged each other. Management responded by closing significant risk positions in the areas that generated losses, pulling back in other strategies, and becoming more selective about capital deployment. Nevertheless, the episode underscores a critical allocator challenge: positions advertised as diversified across "trading," "AI hedge funds," and "tech long/short" strategies may represent a single crowded trade marked in three different places, creating illusion of diversification when correlation becomes unity during market stress.
The losses stemmed from an AI-stock correction that cascaded through multiple layers of Jane Street's portfolio simultaneously. The firm held an LP stake in Situational Awareness, the hedge fund managed by Leopold Aschenbrenner, former OpenAI researcher, which had been a strong performer in the first half of 2026. According to internal notes reviewed by Reuters, the Situational Awareness position "became large by performing well in the first half of the year," then suffered a severe drawdown that left the stake approximately flat on the year, though still profitable over the full holding period. Facing margin pressures, Situational Awareness liquidated the bulk of its public-equity portfolio through a distressed transaction with Citadel Securities.
Jane Street's own technology and semiconductor exposure compounded the impact. The firm's largest memory and semiconductor positions declined roughly 50% during July, while non-AI Asia longs that had generated strong second-quarter returns reversed sharply. The structural failure of the firm's risk management framework became apparent when its standard hedge—short-term put options purchased against tail events—proved ineffective. Because July's AI losses were "relatively spread out throughout the month" rather than concentrated in a single sharp shock, these short-term puts provided minimal protection. The firm noted it "largely lost on the same portfolio of trades that had strong outperformance in the second quarter."
The deeper concern for allocators involves the concentration of the AI trade across seemingly diversified asset classes. Jane Street simultaneously operated as a market-maker providing liquidity in AI-related securities, an LP in an AI-specialist hedge fund with overlapping factor exposure, and a proprietary trader holding correlated long positions. When the reversal occurred, these positions amplified rather than hedged each other. Management responded by closing significant risk positions in the areas that generated losses, pulling back in other strategies, and becoming more selective about capital deployment. Nevertheless, the episode underscores a critical allocator challenge: positions advertised as diversified across "trading," "AI hedge funds," and "tech long/short" strategies may represent a single crowded trade marked in three different places, creating illusion of diversification when correlation becomes unity during market stress.
Go deeper than the headline
Research every firm in this story, track the ones that matter and get the daily briefing — free for 7 days.