Robbie Whelan

AI can now do productive and valuable work.

Tokens are now profitable.

So model makers are in a race to produce more.

In the AI era, compute capacity is revenue and profits.

NVIDIA is a platform for this era.

Yeah, we had another blowout quarter for NVIDIA.

And you would think that when we see results like that, that the stock would really just go through the roof.

but it's very hard to budge NVIDIA's stock.

What we've noticed in the last couple of quarters is that the beat, that is the margin by which NVIDIA's results beat analyst expectations, is almost taken for granted at this point.

It takes a really big beat for NVIDIA's stock to really see any kind of significant movement.

The other thing is that the market is really rapidly changing for AI chips.

We're no longer just training the chat GPT type models that began the AI revolution a couple of years ago.

We're in this new phase where we're trying to monetize everything.

And what that means is the companies that are developing AI tools need AI to be affordable.

What you're seeing is a ton of new entrants into the market of AI chips.

That's not necessarily a problem for Nvidia, but it does change the situation for them a little bit.

Yeah, Dan, I've been thinking about margins a lot, too.

You know, with training in this old era that we were in where NVIDIA was the undisputed top dog in selling these really fancy GPUs, which were really kind of like the Ferraris of computing,

They could charge whatever they wanted.

There was almost no competition.