Hedge funds dumped global technology stocks at one of the fastest paces on record last week, a sharp reversal that hit the market’s most crowded growth theme just as investors started questioning how long the artificial-intelligence buildout can keep outrunning politics and valuations. Goldman Sachs prime-brokerage data through Nov. 6, 2026 showed tech was the most net-sold sector globally and in the U.S., while the selling wave helped drive the largest single-stock net-selling burst in four months.
The move lands in the middle of a violent reset for mega-cap tech and AI-linked names. The Nasdaq 100 fell 6.4% in a single day on Thursday, its biggest one-day drop since March 2020, wiping out about $2.3 trillion in market value. For a sector that had spent much of the year carrying the broader market, the speed of the reversal is now forcing a more basic question: how much of the AI story was built on expectations that can still be financed, approved and executed at the same pace?
The immediate trigger was a reassessment of the data-center buildout, according to the reports. Hedge funds moved after concern that political opposition, with Democrats winning Congress in the midterms, could slow or derail the infrastructure spending underpinning the AI trade. That matters because the current investment case for AI is not just about software demand or chip sales. It also depends on a long pipeline of power, land, permitting, construction and capital spending that must keep moving to justify the market’s most aggressive assumptions.
That is why the selloff reads like more than a simple profit-taking episode. If investors start to believe the physical buildout can be delayed, the entire valuation stack changes. The market no longer has to discount only earnings growth and product cycles. It also has to price political friction, slower capacity additions and a less predictable spending path from the firms funding the AI boom. In a market that has treated infrastructure expansion as a near-certainty, even a modest change in confidence can trigger a rapid de-risking.
The Goldman data suggest that is exactly what happened. Hedge funds sold global technology stocks at one of the fastest paces on record, and the percentage of selling in tech roughly matched Goldman’s records dating to 2016. In other words, this was not just a normal rotation out of one favorite group and into another. It was a broad retreat from the sector most closely tied to AI expectations, with both global and U.S. exposure reduced at the same time.
The selling did not stop at sector-level exposure. It set off the largest single-stock net-selling wave in four months, a sign that the unwind spread into individual names rather than staying confined to index hedges. That matters because the AI trade has been one of the market’s most concentrated. When positioning gets crowded, even a small change in sentiment can cause a large amount of forced adjustment as funds trim winners, hedge beta and reduce exposure to the same names everyone else owns.
The scale of the move also suggests hedge funds were responding to more than one shock at once. The midterm outcome added a political lens to a market already sensitive to the durability of AI-driven growth. At the same time, the Thursday Nasdaq 100 slide made the downside visible in real time. When a benchmark falls that fast, managers do not wait for a cleaner narrative. They cut risk first and sort out the story later. That is how an orderly rotation can turn into a stampede.
For investors trying to read the signal, the key point is that tech was the most net-sold sector globally and in the U.S. That broad footprint suggests the reassessment was not narrowly about one company, one chip supplier or one software subsector. It was a wider rethink of the entire theme. The market had been rewarding almost anything tied to AI capacity, cloud buildout or compute demand. Now the same exposure is being treated as a source of vulnerability if growth assumptions prove too optimistic or too dependent on uninterrupted policy support.
The de-risking showed up in hedge-fund positioning data as well. U.S. fundamental long/short funds cut gross leverage by 8.7 percentage points to 207.5%, the largest weekly drop since March 2020, according to Jin10 citing Goldman. Net leverage fell 3.7 points to 43.3%. Those numbers matter because leverage is the fuel that can magnify both gains and losses. When it comes down quickly, it signals managers are not just changing opinions. They are actively shrinking balance-sheet exposure.
That kind of move usually follows a clear change in confidence. Funds can tolerate volatility when they believe the underlying thesis remains intact. But when a theme starts to look politically exposed, valuation-heavy and technically stretched at the same time, the incentive shifts toward protection. The result is often a faster unwind than fundamental news alone would justify. Leverage data do not tell you what to buy, but they do show how much conviction is left in the market after a shock.
The steep fall in gross leverage also hints that hedge funds may be preparing for more turbulence rather than less. If the AI story is now sensitive to legislative outcomes, infrastructure timing and near-term earnings delivery, managers may prefer to keep more flexibility. That does not mean they have abandoned the trade entirely. It means they are demanding a cleaner setup before they rebuild exposure. In a market still dominated by a handful of giant tech names, that can keep pressure on index performance even if the broader economy remains stable.
The tech reset was not the only notable move in the hedge-fund book. U.S. industrial stocks were sold at the fastest pace since April 2020, according to the Goldman-linked data. That is important because industrials are often tied to the capital-spending side of the economy, including the kind of physical investment that supports data-center construction and related infrastructure. When funds are pulling back from both tech and industrials at the same time, it suggests a broader hesitation about the investment cycle behind AI, not just the software and hardware names front and center in the trade.
Still, the story is not that the AI trade is over. It is that the market is repricing how smooth the road will be. The open question now is whether institutional investors re-add AI-related exposure once they get more clarity on the pace of data-center buildout, the durability of valuations and whether companies can keep delivering earnings at the speed the market has been pricing in. Until then, the sector that led the market higher is also the one most exposed to a hard reset in expectations.
