Hedge funds are taking profit after mouthwatering chip stocks rally By Investing.com

Hedge funds were net sellers of U.S. semiconductors over the past month as chip stocks extended a sharp AI-driven rally, with Goldman’s AI semiconductor basket up more than 50% vs. the S&P 500 this year. Funds cut long exposure rather than adding fresh short bets, while overall exposure to U.S. AI stocks remains near record highs. At the same time, short positioning in broad equity index and ETF hedges has risen to a decade-long peak, signaling portfolio protection rather than a broad pullback from AI.

Analysis

The key signal is not “AI is over,” but that positioning is maturing from chase to harvest. When hedge funds trim the most crowded winner while keeping overall AI exposure near highs, it usually means dispersion is about to rise: the index can still levitate, but the marginal dollar shifts toward picks-and-shovels, software enablers, and less-owned beneficiaries rather than the obvious high-beta chip complex.

The bigger second-order effect is on volatility and factor behavior. Record-short broad equity hedges alongside elevated gross leverage imply managers are funding single-name longs with index protection, which can suppress downside in the market average while creating sharper single-stock air pockets if the AI trade disappoints on earnings or guidance. That setup is bearish for momentum factor purity and supportive of relative-value trades versus the semiconductor basket.

Near term, Nvidia’s print is the catalyst that determines whether this is a healthy pause or the start of a broader de-rating. A clean beat can trigger a reflexive short-covering squeeze in the subsector because positioning is already lighter, but the asymmetric risk is that strong results are met with muted follow-through if hyperscaler capex commentary softens or lead times normalize. Over the next 1-3 months, the more important question is whether AI infrastructure spend broadens enough to offset valuation compression in the names that have already discounted perfection.

The contrarian view is that the market may be underestimating how much of the upside has already been pulled forward into the most obvious semis winners. If funds are selling strength without adding shorts, they are effectively saying upside is now lower quality and more fragile; that favors relative shorts against the basket rather than naked bearishness on AI itself. The cleaner expression is to fade the crowded hardware leg and keep exposure to second-order beneficiaries where earnings revisions can still surprise.

Key Decisions for Investors

• Short SOXX or SMH into post-earnings strength over the next 1-2 weeks; use call spreads if you want defined risk. Thesis: the basket is vulnerable to multiple compression even if Nvidia prints well, because positioning has been reset but expectations remain high.
• Pair trade: long MSFT or AMZN vs short an equal-beta semiconductor ETF basket for 1-3 months. Thesis: hyperscaler AI monetization has more durable cash-flow support than the most crowded chip names; risk/reward improves if capex guidance stays elevated but sentiment rotates.
• Buy near-dated NVDA put spreads into the earnings window only if implied volatility remains below its recent 12-month percentile. This is a volatility expression, not a fundamental short; upside is limited but a guidance miss or cautious capex commentary can reprice the whole complex quickly.
• Overweight less-consensus AI enablers such as ASML or CRDO on a 3-6 month horizon if you want to stay long the theme without owning the most crowded U.S. semis beta. The setup benefits from ongoing AI spend even if U.S. chip multiples de-rate.
• If the market sells off on Nvidia but semis hold relative strength, fade the panic by taking a small tactical long in broad indices hedged with semiconductor shorts. That combination exploits the current hedge-fund posture where index protection is high but conviction shorting is still absent.