Test report DSG-3003 · Rev B · tested November 18, 2026

AI Datacenter InfrastructureDevice under test

Tech overweight falls 25 points in the largest monthly drop since 2006

Bank of America’s November survey shows technology overweight falling from 37% to 12%, while 52% of managers expect an AI hyperscaler to cut capital spending.

Read
4 min
Words
707
Node
45nm
Operator
Marcus Bennett

Spec summary

  1. Technology overweight fell to 12% in November from 37% in October, the largest one-month drop since June 2006.
  2. 52% expect at least one AI hyperscaler to cut capital spending within six months, up from 30% in October.
  3. Cash holdings averaged 3.6%, while the average year-end 2027 S&P 500 forecast was 7,926.
  4. XLK fell 11.02% over the past month and QQQ fell 7.45%, while XLK remained up 13.8% for the year.
  5. The Federal Reserve meets December 8-9, with a rate hike widely expected.
Fund Managers Are Cutting Tech at the Fastest Pace Since 2006
Fig. AFund Managers Are Cutting Tech at the Fastest Pace Since 2006 — LEVEL_uP_PICTURES / pixabay

Marcus Bennett

Technology overweighting fell to 12% in November from 37% in October, a 25-percentage-point drop and the largest one-month decline since June 2006, according to Bank of America’s November Global Fund Manager Survey released on November 17, 2026.

Technology remained a net overweight. The survey records stated allocations rather than confirmed trades, so it does not prove managers sold 25 points of exposure. Between 169 and 193 managers overseeing $461 billion to $526 billion answered the global poll from November 6 to November 12.

Markets react to the survey

Stocks fell on the survey’s release date. The S&P 500 dropped 1.03% to 7,340.53, logged its fourth straight lower close and moved down from 7,619.38 on November 11. The State Street Technology Select Sector SPDR ETF (NYSEARCA:XLK) closed lower at $163.26.

The Invesco QQQ Trust (NASDAQ:QQQ) fell 0.73% to $656.95. Salesforce (NYSE:CRM) dropped 9.45% to $194.52. Both funds rose after hours after NVIDIA reported, with XLK at $166.78 and QQQ at $665.93. Those quotes can change before the open.

AI spending comes under pressure

Now 52% of managers expect at least one AI hyperscaler to cut capital spending within six months, up from 30% in October. No hyperscaler has announced such a cut, leaving the forecast unconfirmed.

A record net 38% of managers say companies overinvest in AI. Only 16% want companies to raise capital spending, the lowest share since May 2020. Managers also named an AI bubble as their biggest tail risk.

Hyperscaler capital spending supplies revenue for chipmakers and the wider AI supply chain. When investors pressure the largest buyers to spend less, they threaten sales for the companies carrying the most weight in XLK and QQQ.

Debt now links the stock and bond risks. In a separate survey question, 38% picked AI hyperscaler bonds as the most likely source of a credit event, ahead of private credit at 30%. AI buildouts increasingly use debt, and higher bond yields would make the spending that supports chip revenue harder to finance.

Low cash leaves room for more selling

Average cash holdings were 3.6%, which Bank of America describes as extremely low. Almost no manager expects a recession. The average forecast for the S&P 500 at the end of 2027 is 7,926.

Market bottoms usually follow rising cash and worsening sentiment. This survey showed the opposite. Managers cut technology exposure while remaining fully invested and optimistic, which leaves them room to sell more.

Michael Hartnett’s team reads the results as investors telling companies to “pause AI capex.” The team also sees the survey as a signal that the AI boom is ending. It said the results do not yet show “enough fear to be a buy signal” and remain tactically risk-off.

Rates add pressure. The Federal Reserve meets on December 8-9, and a rate hike is widely expected. Higher rates would hurt highly valued growth stocks most.

Index concentration keeps the exposure visible

Index funds carry much of this risk. NVIDIA accounted for 14.65% of XLK and 7.6% of QQQ as of June 30. XLK’s next two holdings represented 12.85% and 8.38% of the fund.

XLK fell 11.02% over the past month, while QQQ fell 7.45%. XLK still held a 13.8% gain for the year.

What does this change for XLK and QQQ?

XLK carries the higher risk because its largest positions depend on AI spending that managers now want to reduce. A rate hike could also compress their valuations. The 13.8% year-to-date gain leaves some buffer if the AI trade weakens further, but low cash and limited recession fear leave managers able to sell.

An equal-weight S&P 500 fund spreads exposure, so no single chip supplier drives its results. Companies using AI to cut costs also carry different risks from those building AI infrastructure.

Two events would suggest that selling pressure is easing:

  • The December survey shows cash holdings rising well above 3.6%.
  • XLK holds above $163.26 after the December 8-9 Federal Reserve meeting.

Either outcome would suggest the selling may be done.

via 247wallst.com (Source)

Filed under

  • ai-capital-spending
  • hyperscalers
  • fund-manager-sentiment
  • xlk
  • qqq
Share this article:

More from Marcus Bennett

Marcus Bennett

Show full bio

News editor covering regulation, policy and their impact on hardware layer of artificial intelligence at Die Signal.

681 articles

Same lot · LOT-362E

« Previous articleNext article »