Brian

Claude 5.1 and the previous versions of Gemini all had a similar cadence or way that they interacted with you.

It is not until Gemini 3.3 Pro that I have seen a distinctly different way that the AI is interacting back with me.

And it's across all builds.

It's not because of prompting.

It is because of the way that model is built.

I think that is the next iteration, next version of how you interact with AI.

And I think they did it the right way by probably building a year or two ago and saying, we're going to build this from the ground up as multimodal.

Because what I think you're seeing now is that

This was a smart move for Gemini and Google in terms of how you're going to interact with the Gemini models in all the other applications that Google owns from YouTube all the way over to building slide decks and Google Docs and things like that, as well as many others, I'm sure.

But it's fundamentally different for me.

It's a different experience.

And I think...

I don't really care much about a 5.2 or 5.3 that, you know, get rid of the MDash, although that was great that 5.1 finally did seem to solve it.

I haven't seen MDashes for anything new now for it.

And that's great.

It's a nice thing to see that they could finally solve for that.

The end of the day, though, is 5.1 still acts and reacts like 5.

And as much as people love 4.0, it reacts like 4.0.

I don't have a problem with that per se, but you've shown me something different.

And if other people are seeing what I'm seeing, which I get is sort of the general consensus, I do not hear people talking negatively about Gemini.