Key Takeaways
- Jim Covello from Goldman Sachs urges investors to pivot toward hyperscalers such as Amazon, Microsoft, and Alphabet instead of semiconductor companies
- Cloud computing giants have experienced valuation compression due to investor doubts about AI capital expenditure returns
- Semiconductor stocks have surged almost 150% over the last twelve months via the Philadelphia Semiconductor Index, creating stretched valuations
- Covello identifies two potential winning scenarios for the hyperscaler investment thesis: demonstrated AI ROI or reduced capital spending boosting cash flows
- The primary downside risk involves hyperscalers continuing massive expenditures without delivering measurable returns
Goldman Sachs equity research analyst Jim Covello believes Wall Street has been chasing the wrong segment of the artificial intelligence investment opportunity. According to his latest assessment, cloud computing hyperscalers present superior value compared to semiconductor manufacturers.
Covello, who serves as co-head of equity research at Goldman Sachs while also maintaining coverage of the semiconductor industry, presented his analysis in a client memo distributed Thursday.
The core of his thesis revolves around current market valuations. Major hyperscalers including Amazon, Microsoft, Alphabet, Meta, and Oracle have experienced significant multiple contraction. This reflects growing market skepticism regarding whether these technology giants can generate adequate returns on their substantial artificial intelligence infrastructure investments.
Meanwhile, semiconductor manufacturers have emerged as the preferred AI investment vehicle among market participants. The Philadelphia Semiconductor Index has delivered impressive gains approaching 150% during the past year.
This sustained rally has elevated chip manufacturer valuations well beyond their long-term historical averages. Conversely, hyperscaler stocks continue trading beneath their typical valuation benchmarks.
Dual Pathways to Investment Success
Covello outlined two distinct scenarios under which prioritizing hyperscalers over chipmakers could deliver superior returns.
The first pathway involves hyperscalers beginning to demonstrate meaningful positive returns from their AI investments. Such evidence would alleviate investor apprehension and drive stock appreciation. Semiconductor stocks would face limited upside since markets have already priced in optimistic expectations.
The alternative scenario envisions hyperscalers reducing capital expenditures should return profiles remain disappointing. While this might initially appear negative, Covello contends it could actually support stock valuations through enhanced free cash flow generation. Simultaneously, such spending reductions would negatively impact semiconductor companies dependent on continued hyperscaler demand.
Primary Downside Consideration
Covello identified a middle-path outcome as the principal risk to this investment thesis. Should hyperscalers maintain elevated spending levels without delivering transparent returns, their equity valuations could face sustained pressure.
This same environment would perpetuate robust chip demand, continuing to underpin semiconductor sector performance.
Covello’s analysis arrives amid intensifying investor scrutiny of major technology companies’ data center and AI infrastructure capital allocation strategies.
Amazon, Microsoft, and Alphabet have each announced substantial capital expenditure commitments extending through 2025 and subsequent years.
Meta has similarly accelerated its AI-related investments, prompting questions about the timing and magnitude of revenue conversion.
Oracle has established itself as a significant participant in AI cloud infrastructure, reporting robust demand across its service portfolio.
The semiconductor industry has captured significant benefits from this capital spending cycle, with AI data center chip suppliers reporting healthy order backlogs.
Covello’s recommendation represents a contrarian position given recent market dynamics, though it stems from fundamental valuation analysis relative to historical norms.





