Andy Halliday

And the chip competition that's happening now, for example, Anthropic...

Now intending, not just intending, but in the planning and implementation of a million of the Tensor processing units from Google that would otherwise have been a million of NVIDIA chips somehow.

It's like there's competition now for the chip side of the business, but hey, NVIDIA is fully diversified.

All the way from robotics to text-to-speech, all of that.

They have everything.

Like the Jetson series of compute platforms for autonomous embodied AI, you mean?

It's a multi-stack.

It's a multiple hardware stack system that altogether is much more than just the chip.

It's more than a GPU.

It's multiple components all assembled together to be much more efficient and designed particularly for AI inference.

I didn't read it, and I saw the news item about the neural switch that Anthrop was talking about.

I don't understand exactly whether they've understood where in the deep neural network that function is actually occurring.

And that would give them the ability to contain it, like to try to block it off if they want to preserve alignment, for example.

Or what's the outcome of their having identified the location of that in a deep neural network or in the processing of a deep neural network?

Yeah.

is the import that we now have better transparency of what's going on in the innards.

And now we can control that.

So that's, I placed it as a concept in my mind that I go back and figure out.

But you know, that's what I believe it is, is that they're identifying something that, that gives them more controllability and,

over the outcome of a model as it progresses in its intelligence.