Nvidia is back on top. The chipmaker just took back the crown as the most valuable U.S. company, shares up about 15% year to date after a second-quarter earnings beat and some real momentum behind its next big platform, Vera Rubin. Seeking Alpha has the full breakdown if you want the nitty-gritty.

If you've been reading this site for a while you know I watch this company the way my friends watch reality TV. Not because I'm loaded up on it, I'm not, but because Nvidia is basically the scoreboard for the entire AI spending story I keep writing about. Meta, Alphabet, Alibaba, all of them are shoveling billions at AI compute and most of that money lands in Nvidia's pocket on the way through.

The Q2 numbers beat expectations, and the part that matters most to me is Vera Rubin picking up actual customer traction. That's the successor to Blackwell, and buyers already lining up tells you the demand isn't cooling off. This is the loop that keeps the whole thing spinning: cloud giants spend on chips, the chips power AI services, the services (supposedly) make money, which justifies buying more chips.

Warren and I have been arguing about this stock for months. He thinks the capex cycle has to slow down eventually and he's probably right, cycles always do. My counter is that even if spending flattens, Nvidia's installed base and software moat (CUDA, the thing every AI developer gets trained on, a term I only know because of arguing with him) makes this less of a commodity business than the chipmakers of the past. We're both hedging. He won't buy it, I won't buy it, and we both keep watching.

So should you buy it here? I have no idea, and neither does anyone else yelling about it online. A 15% run on top of an already enormous valuation means a whole lot of future growth is baked into the price already. If AI demand stays this hot the stock probably goes higher. If one big customer blinks and cuts capex you get a 20% haircut in a week. That's not a prediction, that's just how stocks priced for perfection behave.

What I actually do remains boring. I bought my index funds on the first of the month like always and Nvidia is one of the biggest holdings in those funds, so I already have plenty of exposure whether I like it or not. You probably do too if you own any broad market fund. So when people ask me if they're missing the Nvidia boat, my honest answer is they're likely already on it, just at cruise-ship scale instead of speedboat scale.

If you want to make a concentrated bet anyway, size it like a topping, not the meal. Same rule I gave for those energy stocks yesterday. The AI buildout is real, Nvidia's position in it is dominant, and I'd still cap any single stock at a slice of your portfolio you could watch drop 30% without losing sleep. Because at some point it will, and the people who sized it like a meal are the ones who sell at the bottom.