Steve Brown

Well, in my case, there was one very specific one.

So this drug,

was super effective, but it's slowing down a little bit.

And so I'm asking the AI, well, what else don't I know?

And it says, look, it doesn't matter what time of day you take it, just as long as you take that dose with 40 grams of fat.

I'm like, well, the pharmacist said with food, take it with food.

I was doing it with like, you know, a low fat snack in the afternoon.

Well, now I'm like, yeah, now I'm like now I'm now I'm having it with with a steak dinner and, you know, having the ice cream afterwards.

Yeah.

But I mean, but the fact is, it's like like that's just a little piece of information that's.

pretty important that no one told me that.

But how many things like that are there about things in our lifestyle, things that we do, and all that month before we see our doctor, there's a lot of little things we're doing that might have an impact.

So how do you take all of that information and optimize that so you're getting a better response to what you're on?

There are so many variables.

It's hard to imagine

Figuring that all out, all of that out unaided.

if you're shoving everything into the context window, like your whole medical record, and you're overloading the context window, even though, hey, it holds a million tokens, that doesn't mean it can make sense of a million tokens.

So you need to make sure you don't overload it, and you need to focus on the right thing.

So you need to kind of summarize things.

There's a lot of information that might be unnecessary, and there's other information that's important.