A slowdown in the pace of AI development wouldn't necessary dent spending, but there are other reasons semiconductor investors may want to take a measured approach

Fears of an AI slowdown sent shares of the iShares Semiconductor ETF tumbling nearly 6% on Monday.

Semiconductor stocks have become a popular way for investors to gain exposure to artificial intelligence. But now, they've turned into the market's pain trade.

After top industry leaders like Anthropic CEO Dario Amodei called for a technology slowdown over the weekend, investors are calling into question the aggressive growth projections driving the high valuations of chip stocks. Shares of the iShares Semiconductor ETF SOXX fell nearly 6% in Monday trading.

For investors wondering what comes next, Monday's volatility offers some hints about the drivers of the AI trade.

"There's been a lot of spending and a lot of buildout in the AI space...The market has been pretty optimistic about the way it prices the future growth of everything tied to the AI story," Charles Rinehart, chief investment officer of Johnson Investment Counsel, said. "This is a good reminder that things don't go up in a straight line."

Jake Behan, head of capital markets at ETF provider Direxion, pointed out that popular trades can be more susceptible to volatility, especially with some participants looking to trade the news. "It's not surprising to see some air come out of them in a quick way," Behan said regarding semiconductor stocks.

"There's a lot of reasons for traders to take risk off the table," Behan said, with investors wondering if they "still want to be in a crowded chip space with this slowdown question overhanging right now."

Could the recent chatter about an AI slowdown result in a significant reduction in future demand for processing power, or is it just the latest headline causing volatility among chip stocks that have run up a lot this year?

"It's hard to say," according to Rinehart, who pointed out that an AI slowdown might not be feasible in the current political environment, as it could make the U.S. lose geopolitical ground with China.

Bernstein's Stacy Rasgon, a semiconductor analyst at Bernstein, wrote that "recent AI revenue targets from our companies should already be representative of spending plans, and we would be very surprised to see those plans changing materially."

Even if companies slow the pace of AI training an advancement, that won't necessarily lead to a spending slowdown, he said.

The U.S. 10-year Treasury yield briefly surpassed 5% on Monday, a sign that borrowing costs could rise across the economy.

President Donald Trump dismissed the idea of increased AI guardrails in a Truth Social post on Monday. "We are leading China, and all others, and will continue to do so," he wrote.

As a result, investors shouldn't solely fixate on the possibility of an AI slowdown. Beyond debates surrounding regulatory policy, investors should be mindful that volatility is increasing as the AI trade transitions into a new phase - one that's more dependent on debt instead of free cash flow. As a result, macroeconomic headwinds such as rising interest rates will become a bigger factor to the long-term success of the AI trade at large, according to Rinehart.

"We have to be mindful that there's potential for disruptions from other systemic things like interest rates or energy prices that might not, at their surface, tie directly to the AI buildout," Rinehart said. "But as those firms get more dependent on capital markets, something exogenous could create real problems."

Increased volatility and recent headlines don't mean that chip names are suddenly a bad investment. Nancy Tengler, CEO and chief investment officer of Laffer Tengler Investments, sees the selloff as "more likely to be a hiccup" than the end of AI trade.

"The AI genie is out of the bottle," Tengler wrote in a Monday note, highlighting that AI adoption is occurring across diverse industries outside of tech and leading to productivity improvements.

Tengler acknowledged that chip makers and memory providers could be initial losers from increased regulation, but she highlighted that investors are still trying to sort out the implications of an AI slowdown. She thinks that investors should let the short-term volatility settle "and then step in and buy the high-quality names."

Tengler is taking advantage of the weakness in the AI infrastructure trade by adding to stocks like Amazon (AMZN), Micron (MU) and GE Vernova (GEV).

Today's volatility is a reminder for both AI bulls and skeptics to diversify their tech holdings. "If all the spending that we're seeing on AI infrastructure is going to bear fruit and be productive, it's going to have to generate profits outside of just the manufacturers of the equipment," Rinehart said.

However, he flagged that the AI boom has also changed the structure of the stock market. While ETFs and passive investing have historically been a good way to introduce diversification, market concentration around megacap tech names can result in portfolios having more exposure to AI infrastructure names than investors realize.

"More equal-weight approaches or active management might be better compliments for investors looking for diversification," Rinehart said.