Seth Fiegeman

So with training, you want those GPUs the best possible, but with inference, CPUs and some maybe second-tier GPU-type chips could meet the needs.

So maybe that's where we see them play more.

I think it's a little bit of all those.

And I think what we're seeing with OpenAI in particular is that now that the exclusivity arrangement with Microsoft is kind of a done deal, they can tap everyone.

They already have Google as a partner.

They have Amazon.

They have NVIDIA, Broadcom, everyone.

And I think Amazon's ready to play ball in that category.

Yeah, absolutely.

It certainly feels that way.

This is the latest in a series of street high price targets that we've seen on NVIDIA.

Before Loop, we had HSBC, which forecast, I think, 70% or 80% upside.

And before that, we had an additional Loop street high price target.

People keep sort of outdoing each other, trying to estimate how high up this stock is going to go.

So right now, the latest estimate points to about $8.5 trillion

market capitalization.

Like you said, it just hit $5 trillion.

So that points to still a lot more potential upside from here.

And the story hasn't really changed.

People continue to see a lot of long term demand for all kinds of AI related hardware and infrastructure.