Key Takeaways
- • Hedge funds have turned underweight the broader AI trade for the first time since early 2023, according to Morgan Stanley, a sharp reversal after years of heavy AI-linked positioning.
- • Cumulative gross hedge-fund flows have fallen 8% from their June peak, following the longest sustained stretch of hedge-fund de-risking in more than 15 years, per Morgan Stanley Prime Brokerage data.
- • U.S. long-short net leverage rose five percentage points to 51%, but mainly because investors closed short positions; gross leverage fell to 207%, signaling shrinking books rather than rebuilt long exposure.
- • Goldman Sachs Prime Brokerage data showed funds were net buyers of U.S. equities during the Oct. 23-29 period for the first time in four weeks, yet 73% of that buying went into indexes and ETFs, with tech purchases recovering only about 40% of prior selling.
- • Morgan Stanley's prime desk noted depressed net exposure could leave room for more short covering on an eventual turnaround, making the AI trade's lack of crowding a potential risk or opportunity.
Summary
Morgan Stanley data indicates hedge funds have moved underweight the broader artificial-intelligence trade for the first time since early 2023, a major reversal after years of heavy positioning in AI-linked stocks. The shift follows the longest sustained stretch of hedge-fund de-risking in more than 15 years. Morgan Stanley Prime Brokerage figures show cumulative gross hedge-fund flows have fallen 8% from their June peak, as managers repeatedly cut exposure through the summer and October selloffs without rebuilding in between. AI and technology stocks have been central to that retreat, with investors increasingly questioning whether heavy spending on AI infrastructure is generating returns quickly enough. Recent earnings from Alphabet and Amazon added to those concerns, according to Morgan Stanley.
Recent buying appears less bullish than the headline suggests. U.S. long-short net leverage rose five percentage points to 51%, but that increase was driven mainly by investors closing short positions. Gross leverage fell to 207%, indicating hedge funds were shrinking their books rather than aggressively adding long exposure. Goldman Sachs Prime Brokerage data showed a similar pattern: funds became net buyers of U.S. equities during the Oct. 23-29 period for the first time in four weeks, but 73% of that buying went into indexes and ETFs. Technology-sector purchases recovered only about 40% of the selling seen during the previous three weeks, and the Magnificent Seven were net sold in four of five sessions.
The commentary frames the episode as a positioning reset rather than a simple AI-stock correction. Hedge funds have moved to underweight while keeping overall leverage unusually low, which creates two competing implications. The bearish reading is that institutional investors remain worried about AI capital expenditure, free cash flow and whether returns can justify current valuations. The alternative reading is that very light positioning could fuel a rebound if earnings stabilize or sentiment improves, since managers would have room to rebuild exposure or cover additional shorts. Morgan Stanley's prime desk specifically noted that depressed net exposure could leave room for more short covering in an eventual turnaround.
The article identifies several indicators to watch: hyperscaler capex, AI revenue growth, hedge-fund leverage, Magnificent Seven flows, and whether recent buying shifts from short covering to genuine long accumulation. The AI trade is no longer crowded, which may represent either a risk or an eventual opportunity.
Recent buying appears less bullish than the headline suggests. U.S. long-short net leverage rose five percentage points to 51%, but that increase was driven mainly by investors closing short positions. Gross leverage fell to 207%, indicating hedge funds were shrinking their books rather than aggressively adding long exposure. Goldman Sachs Prime Brokerage data showed a similar pattern: funds became net buyers of U.S. equities during the Oct. 23-29 period for the first time in four weeks, but 73% of that buying went into indexes and ETFs. Technology-sector purchases recovered only about 40% of the selling seen during the previous three weeks, and the Magnificent Seven were net sold in four of five sessions.
The commentary frames the episode as a positioning reset rather than a simple AI-stock correction. Hedge funds have moved to underweight while keeping overall leverage unusually low, which creates two competing implications. The bearish reading is that institutional investors remain worried about AI capital expenditure, free cash flow and whether returns can justify current valuations. The alternative reading is that very light positioning could fuel a rebound if earnings stabilize or sentiment improves, since managers would have room to rebuild exposure or cover additional shorts. Morgan Stanley's prime desk specifically noted that depressed net exposure could leave room for more short covering in an eventual turnaround.
The article identifies several indicators to watch: hyperscaler capex, AI revenue growth, hedge-fund leverage, Magnificent Seven flows, and whether recent buying shifts from short covering to genuine long accumulation. The AI trade is no longer crowded, which may represent either a risk or an eventual opportunity.
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.