Excitement about Artificial Intelligence has been a major driver of the S&P 500’s performance this year. The popularity of OpenAI’s ChatGPT spurred a frenzy of investment in AI infrastructure from the largest tech companies, and industries ranging from grocery stores to tractor parts are all discussing how AI will affect their businesses. Despite its prevalence in conversation, AI-derived profits have been scarce, drawing parallels to the 2000’s Dot-com bubble. However, our investment team is optimistic that AI will provide a decades-long tailwind across the market, not just the few businesses that found themselves in the spotlight this year.

Two-thirds of the S&P 500’s year-to-date 13% return came from the technology sector. Leading the pack is NVIDA: the designer of hardware enabling cutting-edge AI, and the recipient of the lion’s share of AI profits so far. Were NVIDIA (ticker NVDA) not included in the S&P 500, the S&P’s return would drop down to about 10.8%. Prior to AI, NVDA was known for designing high-end graphics processing units (GPUs) used in computers and gaming consoles. With 80% market share in GPUs, NVDA was perfectly positioned to capitalize on the unexpected breakthroughs researchers had in using GPUs to power AI tools like ChatGPT.

As Microsoft, Google, Amazon, and others have rushed to invest in new GPUs, NVDA has profited immensely. However, in the long run, the components for technological breakthroughs are eventually commodified. Apple provides a clear example of this in recent memory: the original iPhone had revolutionary hardware, but 15 years later that hardware is neither unique to Apple nor a sufficient competitive advantage. Today, the optimism for Apple’s future is due to the highly profitable services it offers its customers through the ecosystem it built off the iPhone.

While Apple’s ecosystem had to be built from nothing, numerous companies already profitably serve large markets by selling subscriptions to software. Unlike hardware, where the tides of fortune often shift to favor the latest and greatest, the best software companies enjoy durable competitive moats stemming from the enduring need for consistency in a business’s operations.

Take for instance Microsoft and its suite of Office products. While other firms have created excellent competitors to Word, Excel, and Outlook, none have been able to threaten the Office suite’s dominance. As companies compete to sell new AI applications to businesses, Microsoft is singular in its massive paying customer base. It is far more profitable for Microsoft to sell AI enhancements to existing applications than it is for an upstart to bring a new tool to market. Adobe, with its suite of art and design software, is another firm sitting atop a large network of users eager to use the new capabilities of AI in their work. Both firms are currently launching paid upgrades that add elements of AI to software that hundreds of millions of users already rely on daily.

As AI proliferates, eventually a new generation of companies will emerge. Startups like Instagram, Uber, Spotify, and even Candy Crush found success because of the technological leap in mobile phones, but these businesses to date still struggle to profitably outlast their competition. While hardware companies race to commoditize AI components, and millions of fledgling companies try to win the startup lottery with a new and novel idea, we believe that the best place to be positioned is somewhere in the middle. Companies already selling high-margin software to extremely loyal customers have the best opportunity to launch profitable AI products, and offer investors a rational way to benefit from the onset of artificial intelligence without speculating on computer chips and startups.