Goldman warns borrowed stock bets are clustered around AI shares

Goldman Sachs is warning that investors are using more borrowed money to chase gains in U.S. stocks, with those leveraged bets clustered in the artificial intelligence trade. The concentration matters because forced selling by borrowers can amplify market declines if popular positions reverse.

Christian Mueller-Glissmann and colleagues at Goldman Sachs said in a client note that money flowing into U.S. equities this year is running well above its average pace. The team said investors are increasingly using leverage to take part in the stock rally.

Goldman put net margin borrowing at about $1.4 trillion, equal to 1.8% of the total value of U.S. equities. Margin borrowing lets traders increase exposure by borrowing against their own capital, which can raise gains when markets move their way and increase losses when trades turn against them.

Borrowing adds pressure when trades sour

The risk, according to Goldman’s warning as reported by Fortune, is that leveraged investors may have to sell assets to meet margin calls or reduce risk if prices fall. That can add selling pressure beyond the stocks at the center of the original trade.

Goldman also pointed to signs of increased leverage outside the United States. Mueller-Glissmann said margin purchases of Japanese equities have climbed above $30 billion, the highest level since the global financial crisis.

The bank said investor leverage remains heavily tied to the AI ecosystem. Fortune did not identify specific companies in Goldman’s leverage warning, but the note came as Wall Street continues to debate whether the AI boom can justify the spending and valuations attached to the sector.

Big tech has lagged the broader market

Deutsche Bank strategist Jim Reid and his team said the large technology companies known as the Magnificent Seven have trailed the S&P 500 so far this year, according to a chart cited by Fortune. Reid listed several reasons for that underperformance, including crowded positioning in the stocks and concern over whether AI-related capital spending by hyperscalers will produce enough profit.

Reid’s team also pointed to a more hawkish Federal Reserve stance on interest rates and higher shipping costs following the Iran war. Those pressures have weighed on a group of stocks that played a central role in earlier market gains.

Broader markets were softer in early trading cited by Fortune. S&P 500 futures were down 0.28% after the index gained 0.79% the previous day. In Europe, the Stoxx 600 was flat and the U.K.’s FTSE 100 was down 0.25% before lunch.

Asian markets were mixed. South Korea’s KOSPI fell 2.04%, China’s CSI 300 dropped 0.41%, Japan’s Nikkei 225 rose 0.59% and India’s Nifty 50 gained 0.63%, according to Fortune. Brent crude was at $71 a barrel, down from a $74 high the day before, while bitcoin traded at $58,900.

Some analysts remain positive on AI

Wells Fargo analyst Ohsung Kwon remained constructive on AI stocks, according to Fortune. Kwon and his team told clients that reports of delayed IPO plans by OpenAI and Anthropic could support the market because the AI labs may shift attention back toward adoption and because fewer new shares would be added to the market.

Wells Fargo also said lower token prices could lift demand for computing power, extending the cycle while pressuring AI lab margins. That view contrasts with Goldman’s caution on leverage, showing how the AI trade remains both a driver of investor enthusiasm and a source of market risk.

This story draws on original reporting from Fortune.