According to reports, Lisa Shalett, Chief Investment Officer at Morgan Stanley Wealth Management, has stated that investors should exercise caution regarding the semiconductor sector, as increasing signs indicate that chip manufacturers' pricing power is being challenged. She believes that market optimism over investments in artificial intelligence (AI) infrastructure may have already driven chip stocks to excessive gains, while signs of a slowdown in the growth rate of AI capital expenditures are beginning to emerge.
In a media interview on Friday, Shalett noted that the technological architecture of AI data centers is undergoing changes. An increasing number of hyperscale cloud service providers are beginning to design their own, lower-cost proprietary AI chips to reduce reliance on traditional chip suppliers. She pointed out that a typical phenomenon is currently emerging within the AI supply chain: when supply chain bottlenecks allow certain companies to reap excessively high profits, downstream enterprises accelerate the development of lower-cost alternatives.
Currently, some memory chip manufacturers are benefiting from high profits driven by tight supply and demand, but this pricing advantage may not be sustainable in the long term. Shalett made these remarks as South Korean memory chip giant SK Hynix (SKHY.US) officially listed on the Nasdaq on Friday. The company previously raised $26.5 billion through its initial public offering (IPO), setting a record for the largest U.S. IPO by a foreign company. However, its stock price has experienced significant volatility recently, having fallen approximately 26% from its high last month.
Despite this, Shalett believes that market capital for the AI investment theme remains abundant, and semiconductor sector valuations are already significantly elevated. In a report she issued this week, she noted that several indicators, including the Semiconductor ETF and the Philadelphia Semiconductor Index, show the sector is in a state of "clear overbought" conditions. Data indicates that the price-to-earnings ratio of the Philadelphia Semiconductor Index has more than tripled since 2022.
She also mentioned that Meta Platforms' (META.US) recent adjustments to its AI strategy reflect how some tech giants are beginning to reassess the return on investment for hundreds of billions of dollars in AI capital expenditures. Previously, Meta CEO Mark Zuckerberg stated that the company is considering leasing out some of its AI infrastructure to improve asset utilization efficiency and commercial returns.
Shalett believes this indicates that discussions have already begun within tech companies regarding the pace of AI investment build-out, return on investment, and how to achieve commercialization earlier. She anticipates that the trend of slowing growth in AI infrastructure capital expenditures is still in its early stages but warrants ongoing market attention in the future.
Comments