Key Takeaways
- • Hedge funds have raised exposure to artificial intelligence-linked technology hardware to its highest level since Goldman Sachs began tracking the data, reflecting sustained conviction in the AI infrastructure theme despite near-term profit-taking activity.
- • Despite semiconductor and chip-equipment stocks rallying substantially since mid-April, hedge funds have been net-sellers of the subsector over the past month—representing the most net-sold U.S. subsector—as managers take profits and rebalance portfolio risk exposure after significant gains.
- • Goldman Sachs data identifies eight AI-favored holdings among hedge fund portfolios: Nvidia, Microsoft, Meta Platforms, Amazon, Credo Technology, and Adobe, indicating concentrated conviction in AI-enabling hardware and software ecosystems.
- • Hedge fund selling in semiconductors reflects risk-management discipline rather than fundamental skepticism of AI; managers are mechanically reducing sector concentration following price appreciation while maintaining overall AI exposure through diversified positions.
- • Global hedge funds sold tech stocks at their highest pace in nearly eight months as of late February, coinciding with elevated geopolitical tensions, inflationary pressures, and Brent crude oil trading above $100 per barrel.
Summary
Hedge fund managers have markedly increased their exposure to artificial intelligence-related technology hardware to levels not seen since at least 2016, according to proprietary data tracked by Goldman Sachs. This sustained elevation reflects institutional conviction in AI infrastructure as a foundational investment theme, despite recent tactical profit-taking in specific subsectors. The positioning underscores a bifurcated approach: maintaining elevated exposure to AI-enabling technologies while executing disciplined rebalancing in concentrated positions.
Semiconductor and chip-equipment stocks have experienced substantial price appreciation since mid-April, yet hedge funds have become net-sellers in the subsector over the past month, making it the most net-sold U.S. subsector during that period. According to Goldman Sachs data, this selling activity does not represent a fundamental retreat from AI thematic conviction. Rather, it reflects profit-taking after significant rallies and risk-management requirements: as semiconductor stock prices have risen mechanically, their portfolio weight has increased, requiring some funds to trim exposure to maintain target risk parameters. David Lin, commenting on Goldman Sachs research, emphasized that "hedge funds have not been chasing" the rally and noted that this rebalancing is "a reflection of hedge funds taking profits, taking some chips off the table."
Notably, despite recent selling pressure, hedge fund exposure to semiconductors and related equipment remains elevated on a cumulative basis. Since the start of the prior year, semiconductor and chip-maker equipment companies remained among the most net-bought U.S. subsectors cumulatively, demonstrating that tactical selling represents position management within an otherwise overweight allocation rather than a thematic pivot. Goldman Sachs identified eight AI-related stocks favored by hedge fund allocators: Nvidia, Microsoft, Meta Platforms, Amazon, Credo Technology, and Adobe, reflecting a focus on both AI-enabling hardware and software application layers.
Broader portfolio hedging activity emerged in late February, with hedge funds selling tech stocks at their highest pace in nearly eight months, coinciding with elevated geopolitical tensions in the Middle East, persistent inflationary pressures, and Brent crude oil trading above $100 per barrel. This hedging activity, while reflecting near-term risk management, does not signal a fundamental rotation away from AI infrastructure themes but rather represents tactical portfolio protection amid macro uncertainty and elevated bond yields.
Semiconductor and chip-equipment stocks have experienced substantial price appreciation since mid-April, yet hedge funds have become net-sellers in the subsector over the past month, making it the most net-sold U.S. subsector during that period. According to Goldman Sachs data, this selling activity does not represent a fundamental retreat from AI thematic conviction. Rather, it reflects profit-taking after significant rallies and risk-management requirements: as semiconductor stock prices have risen mechanically, their portfolio weight has increased, requiring some funds to trim exposure to maintain target risk parameters. David Lin, commenting on Goldman Sachs research, emphasized that "hedge funds have not been chasing" the rally and noted that this rebalancing is "a reflection of hedge funds taking profits, taking some chips off the table."
Notably, despite recent selling pressure, hedge fund exposure to semiconductors and related equipment remains elevated on a cumulative basis. Since the start of the prior year, semiconductor and chip-maker equipment companies remained among the most net-bought U.S. subsectors cumulatively, demonstrating that tactical selling represents position management within an otherwise overweight allocation rather than a thematic pivot. Goldman Sachs identified eight AI-related stocks favored by hedge fund allocators: Nvidia, Microsoft, Meta Platforms, Amazon, Credo Technology, and Adobe, reflecting a focus on both AI-enabling hardware and software application layers.
Broader portfolio hedging activity emerged in late February, with hedge funds selling tech stocks at their highest pace in nearly eight months, coinciding with elevated geopolitical tensions in the Middle East, persistent inflationary pressures, and Brent crude oil trading above $100 per barrel. This hedging activity, while reflecting near-term risk management, does not signal a fundamental rotation away from AI infrastructure themes but rather represents tactical portfolio protection amid macro uncertainty and elevated bond yields.
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.