So there's NVIDIA, casually reporting 265% year-over-year revenue growth like it's no big deal, and suddenly everyone I know is asking me the same question over brunch: "Is it too late to jump on the AI bandwagon, or am I about to become the cautionary tale at next year's holiday parties?"
It's the million-dollar question, isn't it? (Or perhaps the trillion-dollar question, given the market caps we're discussing.)
I've been covering tech investments since before the dot-com bubble burst—yes, I'm that old—and there's something eerily familiar yet fundamentally different about this moment. Let me explain.
Think about technological investment cycles like a wedding reception. No, really, stick with me here.
First comes the ceremony—intimate, exclusive, with only the closest connections present. For AI, that phase belonged to the researchers, the true believers, the folks who understood transformer models before most of us could spell "neural network." They've been invested since before ChatGPT was a household name.
Then there's dinner service, when the circle widens. The institutional investors arrive, the smart money starts flowing, enterprise adoption begins in earnest. That's... well, that's roughly where we are now.
The dance floor phase? That's when your Uber driver pitches you his AI portfolio. When your aunt starts a sentence with "I heard on TikTok that this AI company..." We're not quite there. Not yet.
Look, what makes this cycle different isn't just the technology itself—though that's revolutionary enough. It's that AI isn't primarily a consumer adoption story. This isn't about getting people to download an app or buy a new gadget. It's about fundamentally rewiring how businesses operate, from the ground up.
That takes time. More importantly, it takes implementation. And implementation is messy, expensive, and stubbornly resistant to quarterly earnings cycles.
So what's a regular investor supposed to do? (Besides time travel to 2020, I mean.)
The "picks and shovels" approach has merit. Companies like AMD, Broadcom, and Taiwan Semiconductor might lack NVIDIA's spotlight, but they're essential players in the ecosystem. I spoke with several fund managers last week who are quietly shifting allocations toward these adjacent beneficiaries.
AI-focused ETFs like BOTZ or ROBO offer another entry point—though be warned, peek under the hood and you'll often find NVIDIA as the largest holding anyway. So much for diversification, right?
The most intriguing opportunities might lie with the cloud titans—Microsoft, Google, Amazon—who are both building AI capabilities and, crucially, already have the customer relationships to monetize them. Then there's the software companies integrating AI into existing products. Think Salesforce, Adobe, ServiceNow. The AI isn't the product; it's the feature that makes the product stickier.
Is it too late to invest? That depends entirely on your timeframe.
If you're hoping for another 150% return in six months... yeah, probably. That particular train has likely left the station, picked up speed, and disappeared around the bend.
But if you're thinking in years rather than quarters? The party's still going.
(Full disclosure: I own shares in several companies mentioned above through index funds, though none individually. This isn't investment advice—if I really knew where stocks were heading, I'd be writing this from my private island.)
The wisest approach might be the least exciting: dollar-cost averaging into the sector over time rather than making a dramatic all-in bet. Boring, I know, but effective.
As my grandfather used to say before making terrible investment decisions himself, "The best time to plant a tree was twenty years ago; the second-best time is now." Just make sure you're not planting in overcrowded soil where everyone else has already staked their claim.
Because there's one thing history teaches us about technological revolutions—timing matters less than time in. And patience... well, that's always been the hardest investment strategy of all.
