AI News: Argon Is Insane and Muse Wonder May Be Benchmaxxed

This was a good week in the world of AI. We got a new frontier model from Google and another one from Meta, two AI companies that for a while now have been pretty behind in the AI race and it looks like they may have caught up, or at least that's what they say. We're

going to see if that's true because I have a ton of cool demos to show you from each model. We also have a crazy story about a spy from inside of OpenAI and overall every AI company is making it easier and easier for us to vibe code apps and games even if you're a complete

beginner with no technical abilities. It's a very exciting time to be into AI. We're going to talk about all of that starting with these new models. So this week we got two big model releases and just in the past few weeks we got a bunch of other big model releases. So it seems like every week now there are new

frontier models being released and before we dive into the specifics of the models that came out this week and my exact thoughts on them I kind of want to share my philosophy, I guess you could call it, when it comes to evaluating models these days because they've all gotten so good. Really, how much of a

difference does a new model actually make anymore? Well, I decide which AI tools I want to use by looking at three things. One, the model itself, two, the app you use to access that model, and then of course the price. So for instance, Opus 5.5 is a model, I access it inside of Claude the app, either the

desktop or phone app, and the price of it is pretty good. I can get through an entire week using Opus 5.5 quite heavily, so it works for my plan. Now people weigh the importance of each of these three factors very differently, but personally I'm starting to feel like the app itself and the price are more