Anthony Pompliano

But Anthropic has underinvested in building out CapEx and Dario has been outspoken about it.

Well, I'm starting to run into some token usage.

And if you go online, you're starting to hear this more and more.

So Mythos is a little scary in the fact that we might already be.

at recursive self-improvement.

I've talked about it here, but this is when the computer's learning on its own, which means all types of things start to enter closer to AGI.

And most of the readout has been that Anthropic has something real.

Because remember, there have been very few releases that come out.

If you can't beat what they have already, then you're not going to release something because that's bad for your stock price.

So you're waiting.

So I think there's been this speculation as to what's happened.

On the source code part,

I've read the VentureBeat article.

This was kind of scary.

Like I made a joke in Boston.

I made a joke on the video I just did today.

I'm like, this is kind of like if Kentucky Fried Chicken released their secret recipe on like a year into there.

There'd be no Kentucky Fried Chicken because there'd be 8 billion of them that had the same recipe.

They have every model provider has nuances of efficiencies, nuances of the way the model works.

And a lot of these were released in the source code, which means if they really did have something,