Goldman Data: Hedge Funds Sell AI Stocks at Second-Fastest Pace in a Year
Goldman Sachs’ data on AI stocks points to a sharp retreat rather than a complete exit. In July 2026, hedge funds’ US technology exposure had fallen roughly 10% in market value since early June 2026. Yet by October 2026, their allocation to select mega-cap stocks had rebounded even as broader tech positioning eased.
Summary
Key takeaways
- AI-linked stock sales reached the second-fastest pace in a year.
- AI leverage declined but stayed above historical averages.
- Magnificent Seven holdings recovered to about 22% of US equity exposure by October 2026.
According to Crypto Briefing, which reported the bank’s prime brokerage figures, Goldman Sachs characterized the selling as a tactical portfolio adjustment, not a wholesale withdrawal from technology. Its strategists also said many investors retained strong conviction in core AI positions.
Goldman Sachs tracks rapid selling of AI stocks
Hedge funds sold AI-linked shares at the second-fastest pace in a year. The retreat extended across hardware, storage and peripherals, IT services, semiconductors and software, rather than concentrating in a single part of the technology sector.
During six of the eight weeks leading up to mid-July 2026, funds were net sellers overall. According to the bank, this pullback marked the largest cut to their combined US tech exposure seen in more than ten years.
Lower exposure, but leverage remained elevated
Both net leverage and AI exposure declined from their peaks in the second quarter of 2026, but remained above historical averages. That left funds with reduced positions without erasing their above-average exposure to the trade.
Goldman Sachs’ AI stock analysis also identified significant reductions in long positions and early signs of capitulation in some holdings. By October 2026, the bank observed historically low net leverage across various strategies.
A difficult July, followed by selective positioning
July 2026 brought hedge funds their worst monthly underperformance against the S&P 500 in more than 20 years, a setback linked to their AI positions. Funds trimmed broad technology and chip exposure while retaining select mega-cap names.
The distinction became more pronounced by October 2026. Even as their broader tech allocations fell, hedge funds’ stakes in the Magnificent Seven climbed back to about 22% of their US equity exposure.
A shift toward perceived quality
The report interpreted the combination of broad selling and renewed mega-cap concentration as a flight toward perceived quality. It suggested that funds saw greater risk in the middle tier of the AI trade than among its largest names.
It also identified three indicators for assessing the adjustment: the pace of selling, the direction of net leverage relative to historical averages, and any move to trim the Magnificent Seven alongside chipmakers and software firms.
The October 2026 positioning showed that the pullback remained selective: broader technology exposure eased while the mega-cap allocation rebounded.
Disclaimer: The content of this article solely reflects the author's opinion and does not represent the platform in any capacity. This article is not intended to serve as a reference for making investment decisions.
You may also like
CoreWeave default probability exceeds 60%! Jensen Huang "puts out the fire": Financing costs are declining, no need for NVIDIA to backstop
After Jensen Huang's speech, CoreWeave's stock price, which had fallen by as much as 9.5% in early trading, turned positive. The company's five-year CDS rose above 1,190 basis points during Thursday’s session, a historical high reflecting the cost of debt default protection, while yields on some of its corporate bonds have climbed above 15% to 16%. The five-year CDS spreads for Oracle, Amazon, Google’s parent company, and Meta are all at historically high levels. Deutsche Bank analysts believe that as the AI capital expenditure cycle is increasingly supported by debt rather than free cash flow, investors must place greater emphasis on signals from the credit market. Overall, while AI-related financing costs are not yet excessively high, worrying signs have begun to emerge.
Bitcoin volatility trends show Bitcoin is less volatile than 57% of S&P 500 stocks
Wall Street is wary of Bessent making another move to suppress yields; key clues may emerge this Friday
This Friday, major primary dealers of government bonds will receive the U.S. Treasury's routine survey before the refinancing announcement is released. Citigroup expects the survey may hint at a reduction in long-term bond issuance, or even the cancellation of 20-year U.S. Treasury bonds. Morgan Stanley emphasizes: "The next two refinancing announcements will be crucial, especially the one in November, which comes right after the midterm elections."
Risk aversion outweighs inflation concerns, US and European government bonds rise across the board, US Treasury yields plunge more than 10 basis points intraday
Investors concerned about rising inflation are shifting their focus to the potential impact of high borrowing costs on economic growth. Some commentators have linked the recent decline in U.S. Treasury yields to surging financing costs for AI-related companies, suggesting that pressure in the technology stocks and AI infrastructure sectors is driving funds toward safe-haven assets such as government bonds.
Crypto prices
More
