Crowded positioning, high ETF leverage, and a fleeing crypto market are setting up a volatile few months for AI stocks
The AI trade has been one of the most crowded rooms on Wall Street, and someone just yelled fire.
Heading into the summer of 2026, investor positioning in AI and semiconductor stocks has reached levels that are starting to make even bullish analysts nervous. High leverage in AI-related ETFs, sharp swings in semiconductor prices, and a broader sense that everyone is already in the pool have converged into a setup that strategists are flagging as a recipe for a rough few months.
Goldman Sachs strategist Shawn Tuteja has warned that markets are underestimating the likelihood of sharp corrections in AI-related stocks, describing the potential fallout as “violent corrections” even within an overall bullish trend.
How we got here #
The S&P 500 notched one of its strongest two-month runs on record heading into late May 2026, with AI stocks serving as the primary engine.
AI hyperscalers raised their collective 2026 capital expenditure plans by more than $200 billion, bringing total projected spending to over $725 billion.
High leverage in AI-focused ETFs compounds this dynamic. Leveraged ETF investors are not long-term holders making a five-year bet on compute infrastructure. They are traders managing short-term positions, and when prices move against them, they sell fast. That selling can cascade through the underlying stocks, amplifying moves that would otherwise be manageable.
Crypto’s rough summer is part of the same story #
Bitcoin lost approximately one-third of its value year-to-date through early June 2026, a level of underperformance that stood in stark contrast to the enthusiasm surrounding AI equities.
Bitcoin ETFs have seen roughly $4 billion in outflows since mid-May 2026, and the timing tracks closely with the acceleration of the AI equity rally.
What this means for investors navigating the next few months #
Tuteja’s warning about violent corrections is worth taking seriously precisely because it comes alongside continued bullishness on the long-term AI thesis. A correction in a bull market is not a contradiction. The question is whether investors who loaded up on leveraged AI exposure are sized appropriately to ride through it, or whether they are the fuel that makes the correction more severe than the underlying fundamentals would justify.
The $725 billion in projected AI infrastructure spending is a real number with real consequences for companies throughout the semiconductor and data center supply chain. The more immediate risk is for anyone who entered the trade late, sized up aggressively, or is using leverage to amplify returns in a market that has already done most of its running.
Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our