The chip sell-off may just be a prime AI buying opportunity

Shares of global chipmakers have taken a sharp fall this week as investors reconsider their bets on winners and losers from the rise of artificial intelligence.

Analysts point to a range of concerns behind the drop, including a potential flood of cheaper Chinese chips, and worries that tech companies investing heavily in AI infrastructure may face higher debt-servicing costs while struggling to generate sustainable revenues.

Seoul-listed shares of chipmaker SK Hynix have slid more than 50 per cent from June highs, and the Philadelphia Semiconductor Index of U.S.-listed chipmakers has shed more than a quarter of its value over the same period. But despite the scale of the declines, some argue it doesn’t affect the underlying case for investing in AI-related companies.

“It’s a multigenerational opportunity,” said Sri Iyer, managing director and head of i3 Investments at Guardian Capital. “The question is about at what price to buy it, not whether to buy it.”

But investors hoping to capitalize on AI opportunities have also had to wrestle with the question of what to buy. While proponents of the AI revolution talk up potential efficiency gains boosting profits at non-tech companies that use AI tools, such as through chatbots, supply-chain optimization and product innovation, it has been difficult to calculate specific benefits beyond chipmakers and hyperscalers.

“The reality right now is we’re still in the very early stages of AI implementation,” Mr. Iyer said. A significant amount of money is going to enablers, companies that build the technology and infrastructure behind AI, while adopters, organizations that have integrated AI tools into core business operations, are struggling in the market, he said.

That could change. In a report this week, Morgan Stanley equity analysts led by U.S. equity strategist Michael Wilson said that the adoption of AI tools “is moving decisively from experimentation to measurable enterprise value,” with 40 per cent of companies that had adopted AI tools citing at least one quantifiable benefit, nearly double the rate a year earlier.

Gains for companies were concentrated in revenue generation, cost savings, capital efficiency and productivity, the report said, with adoption potentially boosting companies’ net margins by 1 percentage point through 2027.

South Korea $2-trillion stock rout breaks records as SK Hynix results disappoint

“The outlook for AI adopters is becoming increasingly compelling, with the benefits of implementation helping to offset concerns around potential disruption,” the analysts wrote, adding that industries frequently viewed as vulnerable to AI disruption, such as transport, software and professional services, are among “the more attractive adopter groups” based on their research.

A new exchange-traded fund launched this week reflects the shift. Mr. Iyer said that about two-thirds of the new Toronto-listed Guardian i3 AI Technology and Innovation Fund is focused on AI enablers, but the rest is allocated to companies adopting AI tools.

The ETF itself is the product of several layers of artificial intelligence. Mr. Iyer said Guardian uses AI agents to extract evidence of a company’s AI use from its official filings and management transcripts, and to determine the importance of AI to the company’s operations.

Guardian then uses AI models to predict factors such as valuation metrics, growth rates, short-term direction and volatility, and to generate recommendations that are then assessed by portfolio managers, who make the final decisions about which companies to include.

In addition to quantifiable elements such as cost savings and revenue generation from AI, Mr. Iyer said AI agents will consider “textual data sets” to analyze newer AI-related public companies that may lack detailed operational and balance-sheet data.

The strategy gives investors an opportunity to gain more nuanced exposure to AI-related companies than following an index or focusing on high-profile tech names, he said.

“This AI universe is not just one big blob. It’s got multiple subcomponents.”

Investors are nevertheless unlikely to enjoy a smooth ride, even if the outlook for AI companies remains positive. Mr. Iyer noted that both adopters and enablers of AI technology have experienced “excessive volatility” and sharp drawdowns this year. That’s despite continued indications of strong profitability in the sector.

“The recent correction has mainly been driven by fear rather than evidence of things starting to go awry,” John Higgins, chief economic adviser for financial markets at Capital Economics, wrote in a research note.

“Analysts’ expectations for earnings per share in the big-tech sectors combined have not yet shown signs of faltering.”

Mr. Higgins said falling share prices despite steady earnings expectations had the potential to create “room for a sizable recovery if sentiment turns.”