95% of Companies See 'Zero Return' on $30B Generative AI Spend

This is why it's so good to sell shovels, so-to-speak.

In this case, that's NVDA

The shovel business is good as long as the gold rush lasts. Once the gold rush is over, you're going to have to deal with a significant decrease in volume, unless you can find other customers.

Crypto's over, gaming isn't a large enough market to fill the hole, the only customers that could fill the demand would be military projects. Considering the arms race with China, and the many military applications of AI, that seems the most likely to me. That's not a pleasant thought, of course.

The alternative is a massive crash of the stock price, and considering the fact that NVIDIA makes up 8% of everyone's favorite index, that's not a very pleasant alternative either.

It seems to me that an ultra-financialized economy has trouble with controlled deceleration, once the hypetrain is on it's full-throttle until you hit a wall.

or steel and wood for making shovels, TSMC

There aren’t enough GPUs for average gamers to buy anything vaguely recent and they would love to be able to. Making the best GPUs on the planet is still huge and the market is quite large. Scalping might finally die at this rate, but NVDA wasn’t making any of the scalping money anyway so who cares? Data centers and gamers still need every GPU NVDA can make.

Oh there's definitely a market but it's not as big or worth as much the AI market. Gamers don't need GB200s or H100s, and AMD beats Nvidia on price in most segments. Nvidia isn't going to die, but gamers won't fill the demand.

Data centers might, but then they'll need something else to compute, and if AI fails to deliver on the big disruptive promises it seems unlikely that other technologies will fill those shoes.

I'm just saying that something big will have to change, either Nvidias story or share price. And the story is most likely to pivot to military applications.