You’re correct–you actually can’t improve the model in one area without changing the characteristics in every other area. It’s almost like the whole thing is just a lot of linear regression…

As in you change one parameter and the whole equation changes or something else? (I'm bad at math!)

Yep exactly!

I want to point out that Zitron is _just a guy_. He’s not a billionaire CEO. He’s not a politician. He’s allowed to say what he wants whether it’s right or wrong, and _he’s just a guy_. He’s not swinging stock values for a group of insiders by tweeting. He’s not using pension funds to pay for a jacked-up IPO. He’s blogging and making a podcast, neither of which is compulsory for you to read.

You don’t have to spend effort proving him wrong. Just don’t read it and move on with your life. Regardless of which “side” of AI you’re on it’s kinda ridiculous how much effort gets spent on screaming gotcha at this one commentator.

He charges people money for a newsletter that reinforces completely detached from reality viewpoints. If he's just making stuff up, that's called grifting, and it's good people call that out.

> if he’s just making stuff up, that’s called grifting

So then we should be calling Dario out every time he opens his mouth, right?

yes, we should also point out when he makes silly predictions that don't come true.

It's so convenient that the US government already classified China as "authoritarian". If they hadn't, Dario would have to say “it’s a risk that other people build models more powerful than us”. I have to wonder what his response would be if for instance a lab in France or Germany came out with an open-weight model this good.

If only there was some kind of markup language for websites where different tags could have different meanings. If only.

Sure, right now it's only available to $200/mo subscribers. It won't stay like that forever...

I like this quite a bit. Thanks for sharing!