Key Takeaways
- • Jordi Visser, president and CIO of Weiss Multi-Strategy Advisers, argues the October 2026 AI-trade drawdown was a hedge-fund deleveraging event rather than a bubble bursting, and that forced selling is largely finished.
- • Goldman Sachs data show hedge-fund net exposure fell to its lowest level outside "Liberation Day" in five years, and Visser calls the Goldman VIP crowded-long basket's 75-day decline the worst since the GFC outside 2008.
- • Visser's recession framework shows no trigger: continuing jobless claims are not rising, S&P 500 year-over-year is massively positive, and corporate profits are up about 20%, though junk spreads have begun widening from all-time tights.
- • Visser cites SpaceX's earnings call as the strongest demand signal he has seen, with Musk saying compute demand is "substantially greater than supply" and J.P. Morgan doubling its SpaceX price target from $240 to $500.
- • Visser dismisses the Bitcoin-to-zero argument as unsupported, citing Adam Back's view that quantum threats to Bitcoin are 20-40 years away, and points to tokenization and AI-driven drug discovery (Eli Lilly's TuneLab) as his preferred themes.
Summary
Jordi Visser, president and chief investment officer of Weiss Multi-Strategy Advisers, contends that the violent selling in October 2026 was a positioning reset in which hedge-fund leverage unwound, not a bubble bursting in AI. He cites Goldman Sachs data showing hedge-fund net exposure at its lowest level outside the "Liberation Day" period in five years, and says the net buying accumulated since 2025 was fully reversed. In his view the deleveraging is "either done or in the seventh or eighth inning," even though markets closed the week at their lows. He distinguishes this hedge-fund deleveraging from economic deleveraging, in which earnings and margins fall first; he sees only the former now. Visser also notes that the October 22 session saw 324 S&P names rise while the index fell 1.36%, the widest breadth-versus-price divergence since 1990 apart from the DeepSeek panic day.
Visser draws a parallel to the August 2007 quant crisis, describing a mechanism of 18 months of crowded accumulation followed by a forced exit in roughly 75 days. He expects a recovery similar to 2007, when the market "almost made a full recovery" after the factor unwind. He also says that, with index hedges no longer offsetting AI concentration, funds have added single-stock shorts, pushing bearish bets near a five-year high.
On recession risk, Visser says his five-component model shows no trigger. Continuing jobless claims show no recession signal, and corporate profits, which he calls his best leading indicator, are up about 20%. Junk spreads hit all-time tights and are beginning to widen, which he identifies as the one component signaling risk.
Visser calls SpaceX's earnings call the most important release he has seen apart from NVIDIA's. Musk said compute demand is "substantially greater than supply" and that he is "fairly confident" of adding 10 gigawatts of compute by 2027. J.P. Morgan doubled its SpaceX price target from $240 to $500 and forecasts roughly $250 billion of 2027 revenue, according to the summary. Visser also argues AI is not a bubble because demand outpaces capacity, citing Eric Schmidt, Sundar Pichai, and David Sacks, and says the credit side of a roughly $10 trillion AI buildout has "barely started."
Visser argues healthcare and life sciences are beginning a major bull market and calls Eli Lilly "the most important AI company in the world" based on its TuneLab platform. On Bitcoin, he dismisses Chamath Palihapitiya's claim that Bitcoin goes to zero, contrasting it with Adam Back's view that quantum threats are 20-40 years away. Visser's own priority is tokenization, which he is pursuing through a tokenized index he is trying to make tradeable, and he has an interest in that view.
Visser draws a parallel to the August 2007 quant crisis, describing a mechanism of 18 months of crowded accumulation followed by a forced exit in roughly 75 days. He expects a recovery similar to 2007, when the market "almost made a full recovery" after the factor unwind. He also says that, with index hedges no longer offsetting AI concentration, funds have added single-stock shorts, pushing bearish bets near a five-year high.
On recession risk, Visser says his five-component model shows no trigger. Continuing jobless claims show no recession signal, and corporate profits, which he calls his best leading indicator, are up about 20%. Junk spreads hit all-time tights and are beginning to widen, which he identifies as the one component signaling risk.
Visser calls SpaceX's earnings call the most important release he has seen apart from NVIDIA's. Musk said compute demand is "substantially greater than supply" and that he is "fairly confident" of adding 10 gigawatts of compute by 2027. J.P. Morgan doubled its SpaceX price target from $240 to $500 and forecasts roughly $250 billion of 2027 revenue, according to the summary. Visser also argues AI is not a bubble because demand outpaces capacity, citing Eric Schmidt, Sundar Pichai, and David Sacks, and says the credit side of a roughly $10 trillion AI buildout has "barely started."
Visser argues healthcare and life sciences are beginning a major bull market and calls Eli Lilly "the most important AI company in the world" based on its TuneLab platform. On Bitcoin, he dismisses Chamath Palihapitiya's claim that Bitcoin goes to zero, contrasting it with Adam Back's view that quantum threats are 20-40 years away. Visser's own priority is tokenization, which he is pursuing through a tokenized index he is trying to make tradeable, and he has an interest in that view.
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.