Consult a healthcare professional if symptoms persist.

This is a huge one for someone who's not done like a telos or something like that.

Just be like, you know, I'm always jumping around.

I think this might be one of the biggest prompts actually that that's like super useful for people is especially people who have been sort of freaking out about AI and they've just done a million projects and they've tried a bunch of product ideas and some of them worked and some of them didn't and they're feeling like massive anxiety.

I think one good prompt would be like, hey, go look at everything that we've done together.

Right.

Look at all of our history.

What would you, would you recommend using like Opus 4.8 to prepare prompts for Fable?

So for our listeners who might not know, we want you to explain what is Lit RPG and what about it makes it feel productive to you?

yeah yeah so lit rpg is essentially like if anyone remembers dragon ball z um if not it's just like a cartoon but the whole thing for the cartoon and actually a whole lot of anime as well is like you have someone who's like humble beginnings or whatever and then they get some sort of powers and then as they proceed through like fights and stuff or challenges they keep escalating in power Right.

And like Dragon Ball Z was one of the things that like made this really evident.

And then you have role playing games, which by that most people mean Dungeons and Dragons, which means you have like dice and you have like skills or whatever.

But it's the same sort of concept, which is you face challenges and then you escalate like you get a skill, your stat goes up or whatever.

But the thing that.

nerds really like about this is that it's a number.

Uh, if you could talk a bit about, about what that is.

... 'cause I think you prob- you probably have the most, uh, the most seat time trying to explain this.

Yeah.

Yeah.

I- I've actually got a way I, I've been thinking about this.

So I love the, the, the, the fact that security is, like, a combination word from Latin.

It's se- and cura, which is, like, without and worry.

So it's, like, the job of security to make the receiver relax and, and not, not panic, right? And being ex-military, I think about, I, I just hate being surprised.

Yep.

Right?

Yeah, there, there's another way to, to, to see this actually be true is that if you look at, like, the giant banks or really advanced industries that have just been really nailing their stuff for decades and decades, and they have, like, all this knowledge built into their DNA, they can actually explain how processes work Right.

And I'm not saying all, but many of these companies who actually can articulate what they're doing, they, they could pick up AI way faster, and I'm seeing them move much faster with AI.

They're more careful with moving with AI because they're also a careful organization, which is why they have everything documented.

But they actually can move fairly quickly because they have something to give to the AI.

You, to your point, Eric, you can't automate processes that you don't have.

[laughs]

I, I, I'm trying to figure out for, for the statement that you made, Raph, what, what is the problem that you think we need to solve? Okay? 'Cause it sounds like you're reacting to, like, the hype of the AI industry, but, but, like, I think that's got its own, like, marketing problems and just people...

And then otherwise, you know, tell me what do you think, what am I doing right or, or wrong?

I mean, I think that's, that's overall correct.

Um, it sounded, I heard a lot of different repos in there.

Um, I think I'm u- trying to, um, be more unified or simplified.

So I have, um, a uni- unsupervised learning, um, unified GitHub, which is, uh, essentially what are we doing? Work.

It's, so the reason I'm using GitHub for this is because it has so many, uh, primitives already built in.

So, uh, you can use issues.

Do you think that, i-- does that all kind of boil down to you to, like, personal idiosyncratic preference, or do you see structural or performance reasons to create multiple different agents with different personas?

It's like the atomic bomb.

Yeah.

Yeah, and the thing I'm actually worried about, and this is...

One, I'm worried about an actual event, but there's, um, smaller things that could go really, really crazy and just, like, change everything overnight.

Somebody could have an open source model or claim they, they got it from one of the main models, but there could be...

Let's say there's, uh, you know, heaven forbid, forbid some sort of, like, attack.

Um, let's say an entire apartment building smells a smell, and there's a 911 call, and everyone gets flooded out of the building.

What does it mean in this context?