The market is no longer debating whether AI demand is real. The harder question is which
companies can turn that demand into lasting earnings and cash flow.
AI remains the biggest investment theme in technology, but the gains have not been evenly shared. So far, the most obvious winners have been the companies supplying the hardware needed to build out capacity: processors, high-bandwidth memory, networking equipment, advanced packaging, power systems, and cooling. Now the focus is beginning to shift. Investors want to know which companies can turn AI adoption into recurring revenue, better margins, and returns that justify the enormous sums being spent.
See more: The AI Sell-Off and Portfolio Implications
Why Semiconductors Still Have the Edge
Semiconductors remain the clearest near-term beneficiaries because they are paid as capacity is built. Hyperscalers are still investing heavily in computing power, data centers, and the equipment that supports them. At the same time, shortages in memory, networking, advanced packaging, power, and cooling are keeping demand visible and, in some areas, supporting pricing.
That helps explain why earnings estimates for semiconductor companies have generally moved higher while software estimates have been more mixed. The connection between spending and revenue is straightforward. Before an AI system can produce commercial results, someone has to buy the processors, memory, networking, power equipment, and cooling. Hardware suppliers get paid during the buildout; software vendors often have to wait for customers to prove the return.
Software Has More to Prove
Software is a more complicated story. AI tools are moving out of the pilot stage and into everyday use, but not every vendor will benefit in the same way. Companies with proprietary data, control of important workflows, or a strong position in security have a better chance of holding on to the value they create. Others may find that automation reduces seat counts, weakens pricing, or requires heavy investment before new revenue shows up.
Earnings reports are starting to separate the leaders from the rest. The market is rewarding companies that can point to larger contracts, growing backlogs, real production deployments, higher usage, or better margins tied to AI. A good story is no longer enough. The stronger software companies are showing that AI can make their products more useful and their customer relationships stickier. Weaker businesses remain exposed to competition and pricing pressure.
Follow the Returns, Not Just the Spending
The biggest risk may not be that AI fails. It may be that AI works, but the profits land in fewer places than investors expect. Chipmakers sell the components. Hyperscalers fund the infrastructure. Software companies package new capabilities. Their customers hope to gain productivity. Everyone expects a return, but competition and falling computing costs will ultimately decide who keeps the economics.
That makes free cash flow, capital intensity, utilization, and pricing power more important than ever. Valuation matters, too. A company can post strong results and still disappoint if expectations have run too far ahead. It is not enough to own a business simply because it has AI exposure. The real work is judging how large the opportunity is, how long it can last, and how much of the revenue will become profit.
What This Means for Portfolios
The better opportunities are likely to be companies with visible demand, sound economics, and earnings that support the story. In semiconductors, that includes businesses tied to compute, memory, networking, advanced packaging, and power. The trade-off is that expectations are already high, and supply will not stay tight forever. In software, stock selection matters more than broad exposure. Proprietary data, control of workflows, security, switching costs, and proof of AI revenue are the qualities to watch.
There may also be opportunities outside the most familiar technology names. AI data centers require far more electricity, transmission capacity, cooling, and specialized equipment. Companies that enable that expansion could benefit from the same spending cycle without depending on a handful of mega-cap stocks.
The Bottom Line
The AI trade is entering a more demanding stage. Semiconductor companies still have the advantage because they earn revenue directly from the buildout and continue to benefit from tight capacity. Software can produce meaningful winners, but the burden of proof is higher. The question is no longer whether AI will matter. It is who will make money from it—and whether today’s share prices already assume too much.
Authored by Kimberly Woody
For more news, information, and analysis, visit the ETF Strategist Content Hub.
This report has been prepared for informational purposes only. It may include content generated or assisted by artificial intelligence (AI) tools. While the information presented is based on sources believed to be reliable, it should not be construed as personalized investment advice, considered a recommendation or solicitation for the purchase or sale of any security or strategy. Strategy Holdings, Attributions and/or Sectors mentioned are as of the date indicated, subject to change, and should not be relied upon as current. This does not constitute legal or professional advice and is not tailored to the investment needs of any specific investor. Registration of an investment adviser does not imply any certain level of skill or training. Due to rapidly changing market conditions and the complexity of investment decisions, supplemental information may be required to make informed investment decisions, based on your individual investment objectives and suitability specifications. Investors should seek tailored advice and should understand that statements regarding future prospects of the financial market may not be realized, as past performance does not guarantee and/or is not indicative of future results. Content may not be reproduced, distributed, or transmitted in whole or in part by any means without written permission from Globalt. Regarding permission, as well as to receive a copy of Globalt’s Form ADV Part 2 and Part 3, contact Globalt’s Chief Compliance Officer, 3200 Windy Hill Road SE, Suite 1550E, Atlanta GA 30339. You can obtain more information about Globalt Investments and its advisers via the Internet at adviserinfo.sec.gov, sponsored by the U.S. Securities and Exchange Commission. The opinions and some comments contained herein reflect the judgment of the author, as of the date noted.
Globalt Investments LLC (“Globalt” or the “Firm”) was founded in 1990. It has been registered with the SEC as
an Investment Adviser pursuant to the Investment Advisers Act of 1940 since 1991. Effective October 1, 2023, Globalt is a limited liability company owned by the employees and succeeding the “Globalt Investments” which had been a separately identifiable division of Synovus Trust Co. N.A. (its former affiliate since 2002). Globalt is no longer affiliated with Synovus. Globalt’s registration with the SEC does not imply any level of skill or training and should not be mistaken for an endorsement.
A message from Advisor Perspectives and VettaFi: Discover something new! Click here to register for our upcoming webcasts.
More Innovative ETFs Topics >
