Joanne Feeney

You're designing it years out.

And you have to know the supply will be there if you commit to a certain manufacturing partner.

So it's a real chicken and egg problem.

They might have to bet on making it happen and being able to deliver the customers by getting that process up and running and showing that they can get yield up in volume, even in advance of locking down some customers.

So it's almost a bet the company kind of problem.

And that's why, again, as an investor...

We're not enthusiastic about Intel.

The risk is simply too high.

We don't find the current valuation compelling given that risk.

We think there are better places to be, and so we're going to stick with where we are, for example.

And AMD, by the way, that recent result pretty clearly indicated that AMD is continuing to gain share against Intel, likely both in server CPU and in PC CPU.

Well, they seem to be doing a really excellent job of that.

And what's interesting about what Matt just said is they're going to let customer demand drive how they build out these different capabilities for AI applications, which tells us that there are going to be different models purposed to different types of workloads.

So everybody's thinking, oh, NVIDIA is the market-dominant player right now.

They have massive market share.

And they're treating it like it's one sort of big thing that's homogenous.

And I think what people are going to realize over time is that there are going to be different types of models for different workloads.

And different chips will be useful for developing such different models and for running those models, for doing the inference.

So I think there's a lot of room for different chip players here.

And we should expect more of the in-house chips like what Amazon does in collaboration with Marvell or what Google does in collaboration with Broadcom to continue to rise in prominence, as we've been seeing in the recent news coverage.