Key Takeaways
- • Goldman Sachs' AI semiconductor basket has outperformed the S&P 500 by more than 50% year-to-date, but hedge funds are now trimming positions after the sector's near-vertical ascent, moving semiconductors to a net-sold position for the year.
- • South Korea's KOSPI index, a key barometer of global AI infrastructure demand, surged beyond 80% year-to-date before retreating sharply as SK Hynix and Samsung each plunged approximately 12%, triggering broad-based semiconductor sell-offs across Asia.
- • The Philadelphia Semiconductor Index fell around 5% as major U.S. chip stocks including Nvidia, Broadcom, and Micron Technology declined, signaling concerns about AI demand saturation and elevated tech valuations.
- • Hedge funds are simultaneously increasing short positions in broad equity indices and ETFs to hedge wider market risks, with those positions now at decade-long peaks, while maintaining overall gross leverage at five-year highs despite stable net leverage.
- • Profit-taking in semiconductors contrasts with sustained commitment to AI-related investments overall, as funds distinguish between sector-specific valuations and the longer-term AI investment thesis.
Summary
Hedge funds have begun systematically reducing exposure to U.S. semiconductor stocks following an extraordinary rally that saw Goldman Sachs' AI semiconductor basket outperform the S&P 500 by more than 50% year-to-date. According to Goldman Sachs' prime brokerage desk data cited by Bloomberg, semiconductors and semiconductor equipment have become the most heavily net-sold U.S. subsector over the past month, with funds reducing long positions rather than initiating fresh short bets. The sector has shifted to a net-sold position for the full year despite its exceptional earlier performance.
The profit-taking reflects portfolio rebalancing rather than abandonment of the AI investment thesis. Goldman's analysis indicates that overall exposure to U.S. artificial intelligence stocks tracked by its technology, media, and telecommunications basket remains near record highs. However, funds have simultaneously increased short positions in broad equity indices and exchange-traded fund instruments as hedges against wider market risks, with those positions reaching decade-long peaks. This defensive positioning suggests mounting caution about broader equity valuations even as conviction in AI infrastructure remains intact.
The semiconductor retreat has been particularly acute in Asia, where South Korea's KOSPI index—closely monitored as a barometer of global appetite for AI infrastructure—briefly surpassed 8,000 points in mid-May with a year-to-date advance exceeding 80% before retreating sharply. SK Hynix and Samsung each declined approximately 12%, while the Philadelphia Semiconductor Index fell around 5% as major U.S. chip manufacturers including Nvidia, Broadcom, and Micron Technology all traded lower. The declines reflect concerns about AI peak demand, elevated tech valuations, and questions about whether the current buildout costs for AI infrastructure justify current equity pricing.
Notably, Goldman's prime desk characterized these moves as disciplined portfolio management tied to valuations rather than a loss of conviction in AI's secular growth story. Overall gross leverage among hedge funds has risen to a five-year high this month, though net leverage has remained relatively stable—a configuration described as inconsistent with the retail euphoria currently visible in equity markets. This setup suggests professional investors are taking profits on outperformers while maintaining structural hedges and net positioning that reflects underlying market uncertainty.
The profit-taking reflects portfolio rebalancing rather than abandonment of the AI investment thesis. Goldman's analysis indicates that overall exposure to U.S. artificial intelligence stocks tracked by its technology, media, and telecommunications basket remains near record highs. However, funds have simultaneously increased short positions in broad equity indices and exchange-traded fund instruments as hedges against wider market risks, with those positions reaching decade-long peaks. This defensive positioning suggests mounting caution about broader equity valuations even as conviction in AI infrastructure remains intact.
The semiconductor retreat has been particularly acute in Asia, where South Korea's KOSPI index—closely monitored as a barometer of global appetite for AI infrastructure—briefly surpassed 8,000 points in mid-May with a year-to-date advance exceeding 80% before retreating sharply. SK Hynix and Samsung each declined approximately 12%, while the Philadelphia Semiconductor Index fell around 5% as major U.S. chip manufacturers including Nvidia, Broadcom, and Micron Technology all traded lower. The declines reflect concerns about AI peak demand, elevated tech valuations, and questions about whether the current buildout costs for AI infrastructure justify current equity pricing.
Notably, Goldman's prime desk characterized these moves as disciplined portfolio management tied to valuations rather than a loss of conviction in AI's secular growth story. Overall gross leverage among hedge funds has risen to a five-year high this month, though net leverage has remained relatively stable—a configuration described as inconsistent with the retail euphoria currently visible in equity markets. This setup suggests professional investors are taking profits on outperformers while maintaining structural hedges and net positioning that reflects underlying market uncertainty.
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.