If we're thinking about where does the future need to be and where does the frontier need to be, what's that have to look like? Is that a problem of ambition? Is that a problem of know-how? Where, where should it be?

I-I mean, look, I think first of all, I think we owe it to ourselves, but also just like practically, we will be forced to retain the capacity for clinical trials because it turns out That's the end of all of these journeys, right? At, at, at the end of the day, you gotta put patients and humans, you've gotta do it here, you've gotta do it in studies that the FDA will expect.

I accept, I completely agree that data coming into the US has to be held to the same standards wherever it's generated.

That's absolutely true.

Uh, and I think that's an area where we just, we need to be great clinicians, we need to be great clinical scientists.

In terms of, like, other than that, look, I think, uh, that is, I'm sure, partially an industrial policy and industrial capacity question, but it's in part, and this is, like, an annoying glib comment, it's, like, about the spirit of what it means to be an American.

Like, why is it that Google is an American company and then also OpenAI is an American company and that GM and Ford were American companies? Like, why is it that for many, many decades, the first of every one of, like, among the greatest innovations to have come out of whatever technology sector have come from here? I do think that is something that we should not take for granted, and it's something we've gotta figure out how to bottle and do, but I don't think saying, "Well, the last one of those was antibody engineering, so therefore we need to crystallize antibody engineering and be the best antibody engineers forever because that's what worked last time," I don't think that's the answer.

Yeah.

Um, what, what's, what's changing in China?

Yeah.

I mean, I'm on the record in a bunch of different settings as saying I think that China is a little bit of a smokescreen or a red herring for other dynamics that are changing pretty quickly on the technology side around the commodification of antibody development, around certain kinds of even small molecule chemistry, either with or without AI getting easier and more reproducible, and about that stuff getting pushed to places that can do it cheaper, places that can do it faster.

Obviously, China, the current leader there, although I think, you know, to the question of like politically, you know, what, what would happen if we closed off innovation from China, I think it would just happen somewhere else.

B-but, um, look at-- I, I think it's pretty remarkable the speed with which we can now move things.

We collectively, globally, can move things from an idea to an experiment.

Uh, you know, a-and I think it's, uh, it's a great thing that there's more novel targets coming, more no-- I, I don't, I don't-- not particularly biased toward the novel targets.

And do you really need outcomes? And then I'd love to, if you can comment, uh, about the exciting work that Coursera is doing too.

Right

... there's a lot to do.

A- and there's a whole team of people, not me specifically, working on everything from hiring field medical professionals, to building a patient support organization, to figuring out how the pills are gonna get from the manufacturer to the warehouse to the pharmacy to the patient.

Uh, and all of that work is ongoing.

Uh, and it's all-consuming.

It's an incredible effort that takes a village and, uh, I, I believe in the team that we have doing it, but it's a lot of work.

Um- I think one of the great things about this moment in time for us is we do get to stand...

Um, and then, like, how, how are you differentiating to show that the outcome matters in your phase three data versus, you know, what a pure JAK1 or pure TYK2 inhibitor could do?

Hmm.

Not because I don't think it'll matter, and I'll come back to that in a second, but because it feels like the fruit at the very top of the tree.

It feels like of the places that AI will matter in healthcare and in biotech, solving problems of chemistry and biology, where we literally don't have the data required to train the models in some cases, fine, great.

Let's generate that data.

Let's train those models.

Those are hard problems.

And if we can make little bits and pieces of progress there, that's great.

So can you tell me about kind of like the milestones that are coming up?