In an AI Arms Race, Investors Should Focus on Profit Potential

As the AI halo begins to fade, equity investors are seeking companies that can profit from—and not just pontificate about—artificial intelligence.

Artificial intelligence (AI) is everywhere these days. Not surprisingly, companies have tried to pair up with this fashionable concept in any way possible. But we may be seeing the beginning of a rotation out of AI stocks that could ultimately separate the players from the pretenders.

Investors must now decide if this is a momentary blip in the AI story or a rotation toward broader market participation. We believe many high-quality companies that have been left behind in the two-year technology boom could be rewarded if a more sober view of AI-driven shares takes root.

As the Pendulum Swings

There’s no denying the appeal of AI, which promises to revolutionize how companies do business by automating routine tasks that would otherwise consume valuable human resources. Through the first half of the year, it seemed that companies referencing AI on earnings calls or in conference presentations were rewarded with consistent share-price bumps. The result was a valuation halo around anything related to AI, while the rest of the market saw relative valuations wither.

But in July, the pendulum began to swing the other way. It started on July 11, when small-cap stocks staged a meaningful rally, while mega-cap technology valuations faltered. That was followed by another rotation in late July and early August when many tech companies reported earnings. These momentum swings at times involved indiscriminate selling of tangential AI names with otherwise well-diversified business models—creating buying opportunities in the process.

Today’s Profits or Tomorrow’s Promise?

In our view, there’s a distinct difference between companies benefiting from AI today and those hoping to make AI a relevant part of their future business plans. Chip companies such as NVIDIA, which manufacture the GPUs that power machine learning, clearly fall into the first bucket. Similarly, tech companies like Microsoft that are seeing incremental cloud growth driven in part by AI spending could also be near-term beneficiaries. However, many other companies fall into the second bucket—in particular, those planning for AI but not yet profiting from it. It’s these firms that the market has begun to view with a more skeptical eye.

Investors are paying especially close attention to return on AI investment (ROAI)—made all the more important by massive capital spending among technology’s reigning kings. Microsoft is expected to spend roughly $73 billion in the calendar year 2024, followed by Amazon (nearly $70 billion), Alphabet Inc. ($50 billion) and Meta Platforms (just shy of $40 billion). While not all the spending is on AI, the aggregate outlay of about $230 billion in one year is eye-popping, to be sure, and a huge change from the roughly $100 billion spent in 2020 (Display).

Recent commentary by technology-company management teams is revealing. One common theme is the view that underinvesting in AI is a greater risk than overinvesting. As a result, we’re seeing an emerging arms race where no company believes it can afford to let competitors get ahead of it on a new technology.

But this in no way answers the question of how these companies are going to earn a return on $230 billion worth of capital spending in 2024, and likely even higher spending in 2025 and beyond. Spending on AI could benefit each company’s existing business in varied ways. But in our view, these same firms must realize higher incremental revenues to justify all the spending. Whether that will happen remains to be seen.

Language Models Could Drive ROAI

Revenue growth could come in the form of emerging language models. OpenAI, Anthropic, Llama, Gemini and Mistral AI are among the current leaders. OpenAI appears to have a slight edge over its competition, but all the major models are improving their capabilities. How these models will be monetized is so far unclear. We also don’t yet know how commoditized language models will eventually become, which will also play into profitability.

History provides some clues. New technology has often produced “winner take most” outcomes. We’ve seen this not only with internet search, but also with social media, e-commerce and—initially, at least—the cloud. Expecting all the various language models to drive profits may be optimistic. And if owning a language model becomes table stakes for retaining existing business, the “R” in ROAI may be elusive for many companies in the AI space.

Not All AI Investments Are Alike

Despite all the caveats about profitability potential, we acknowledge that it’s rarely been a good idea to bet against huge technology companies over the last 20 years. At the same time, however, we believe investors must be realistic about future returns given the mega-caps’ astronomical level of investment.

As AI evolves, it will be important for investors to stay selective, particularly since owning the entire AI space—as many passive index investors do—is increasingly risky, in our view. While the high-flying mega-caps dominate the spotlight today, some of the greatest long-term opportunities could come from smaller companies that can generate sustainable revenue streams, and not just wishful thinking.