The jury is in for the first half of 2024 and the market did just fine, actually better than fine given the nastiness here and abroad. The S&P 500 was +14.5 percent through June and the tech-laden NASDAQ Composite was +18.1 percent. It continues to be a tech world and we’re just living in it; and AI (artificial intelligence) is still driving the sector.
Luckily, I’ve been focused on tech investing since 1992, so we’re celebrating on James Island at the Meeks. More importantly, my clients are too. We’re +59 percent thus far in 2024 and +78 percent over the past twelve months. But what will I do next? Is the AI Bubble bursting, or has it already popped?
Remember the Dotcom Bubble? After it burst in 2000, the NASDAQ plunged -78 percent before troughing in 2002. The index didn’t recover its highwater mark until 2015. The Wall Street Journal published an article about me then. They were gracious but others weren’t. The other marquee tech investors in that era, who weren’t as smart as they claimed to be when the Bubble was inflating, but weren’t as dumb as they looked when it popped, refused to speak with the journalist, or they were nowhere to be found when he called. Therefore, maybe by default he spoke with me, the only one not in Dotbomb witness protection.
I’ve done well focusing on AI infrastructure plays. These are mostly key semiconductor, computer server, and networking companies whose gear is necessary to “train” large language models (LLMs). AI’s next stage is “inference” where vendors look to put this data and computer horsepower to work on something useful. Given the hundreds of billions that are being spent on the front-end, the hope is that you, me, and the companies we work for go all-in on AI so that we buy their stuff…lots of it.
AI investors should remember the lesson of the California Gold Rush (1848-1855). Those that made a fortune sold the picks and shovels to the gold miners most of whom never found a nugget. I’m confident that heavy spending on AI gear will continue for at least another year or so and that will lead to further gains in the shares of semiconductor companies like NVIDIA (NVDA), Advanced Micro Devices (AMD), Broadcom (AVGO), Micron (MU), and Taiwan Semiconductor (TSM). I also see more money to be made in data server stocks like Dell (DELL), Super Micro (SMCI), and Hewlett Packard Enterprise (HPE). Last, I think that Microsoft (MSFT), Alphabet (GOOGL), Meta (META), Arista Networks (ANET), and Amazon (AMZN) should continue to be key AI beneficiaries.
Why? First, they control the data that’s fed into LLMs. The big winner in almost every tech trend is the firm that controls the data. Second, only these tech titans can afford to spend the necessary money to build AI arsenals. “The race is not always to the swift, nor the battle to the strong, but that's the way to bet.” Newspaperman Damon Runyon (1880-1946) wrote (or plagiarized) this, but it describes why Big Tech is, well, Big Tech.
Yes, I’ve profited from AI infrastructure bets and I’m sticking with them because while the bears believe that someone is about to swipe the punchbowl, I bet that the party rages on. My other contrarian call is that I’m not sold like the bulls are that most AI miners will find gold, or at least enough of it to make their almost unprecedented AI infrastructure spending worthwhile.
Here’s a test case that’s coming this Christmas: Apple (AAPL), which hasn’t been on my tech buy list for quite a while, which had been a good call until recently, will add AI to its high-end iPhones sold this holiday season. It’ll just be a feature on their phones. You and I won’t pay extra for it. Apple and its cheerleaders are betting the ranch that “AI on board” will lead to an iPhone upgrade Super Cycle. Will you dump your phone to buy the most expensive one -- $1,199 for the iPhone 15 Pro Max, but the iPhone 16 series will be introduced in September -- so you can use its AI feature beginning in January? I won’t. Time will tell for the masses.
But know this: Features typically aren’t money makers. They quickly become commoditized. Usually, products, and better yet, product platforms, have the greatest chance to succeed. Also, too many investors assume that their favorite companies in almost every economic sector will be turbo-charged by AI revenue and/or cost efficiencies. I bet, just like in the Net Era, that several firms will be disintermediated by AI and will become obsolete and may even go bust.
I’m looking for important AI “use cases” before I invest in the miners. For now, I’m sticking with the picks and shovels sellers. When I see AI gold, and only when I do, I may change my mind. Again, I’m a contrarian skeptic about AI’s economics once it’s ubiquitous. There’s a stark difference between popularity/ubiquity and profits. For most companies, if AI is just a feature, shareholders could rue the day that their companies flushed a lot of money down the AI drain.